Podcasts about sandboxes

In this episode, Alex Jung, cofounder and CPO of Unikraft, joins us at Heavybit DevGuild. Alex explains how Unikraft went from an academic unikernel project to a cloud platform for running highly efficient workloads. We talk about why unikernels can make applications faster, cheaper, and more secure, how AI agents are driving demand for sandboxed compute, and why the next wave of software generation may require a much more efficient cloud infrastructure layer.Links:• Alex Jung's LinkedIn• Alex Jung on X • Unikraft• Unikraft on GitHub • Unikraft open source project 

AI Safety Basics, Exploit Gym, and a 438km Tesla FSD Trip — Plus Kimi K3 and Agentic Tools With John away, Jim and Marcel discuss recent AI security concerns, arguing that sensational claims about an OpenAI model "escaping" distract from basic safeguards like network segmentation and monitoring, especially when testing hacking capability via the Exploit Gym benchmark in a sandboxed, virtualized environment. They pivot to self-driving, with Marcel describing a new rear-wheel-drive Tesla Model Y that drove 438 km to Gananoque and back on supervised Full Self-Driving, including automated routing to Superchargers and self-parking, and they note winter and sensor considerations plus lower fueling costs. The conversation shifts to enterprise AI costs and Chinese models like Kimi K3, Qwen, GLM, and DeepSeek, emphasizing experimentation, data governance, and that huge models require data centers. They highlight agentic systems that can install and run software, memory tools like Mnemosyne, and reflect on how AI could free time for higher-level thinking. 00:00 John Is Away 00:47 Writing And AI Escapes 02:24 Movie Hacking Myths 04:45 Robot Ducks And Chicks 06:04 Exploit Gym Explained 10:36 Agents And Emergence 12:45 Security Lessons Learned 17:05 Self Driving In China 18:35 Tesla Road Trip FSD 25:45 Winter Range And Sensors 28:48 EV Range Reality Check 29:22 Charging Costs and Solar 30:17 Why CIOs Fear AI Costs 32:29 Chinese Models and Multimodal 33:21 Open Source Catch and Sandboxes 35:30 Picking the Right Model Mix 37:33 Innovation vs Reliability Lessons 41:18 Foreign Car Analogy for AI 45:41 What to Use and Pay For 47:33 Voice Mode Gets Real 49:40 Agents That Do the Work 54:20 Memory Plugins and Skill Atrophy 57:59 Habits, Choice, and Closing Thoughts

AI Agents sind mehr als der nächste Technologie-Hype. Richtig eingesetzt können sie recherchieren, analysieren, Entscheidungen vorbereiten und komplexe Aufgaben weitgehend selbstständig bearbeiten. In dieser Folge sprechen Andreas Wiener, Christian und Nico darüber, wo agentische KI heute bereits echten Mehrwert liefert: im Vertrieb, Einkauf, Reporting, Kundenservice und in internen Supportprozessen. An konkreten Beispielen zeigen sie, wie Agenten Informationen zusammenführen, Auffälligkeiten erkennen, Handlungsempfehlungen entwickeln und Mitarbeitende von zeitintensiver Routinearbeit entlasten. Gleichzeitig wird klar: Nicht jeder Prozess braucht sofort ein autonomes Agentensystem. Entscheidend ist, den Use Case sauber zu verstehen und anschließend die passende Technologie auszuwählen – vom klassischen Workflow über Machine Learning bis zum leistungsfähigen AI Agent. Es geht um Datenzugriffe, Sandboxes, Governance, Wirtschaftlichkeit und die richtige Balance zwischen Autonomie und menschlicher Kontrolle. Denn das größte Potenzial entsteht nicht dadurch, Menschen zu ersetzen. Es entsteht, wenn Agenten fähige Mitarbeitende schneller, produktiver und wirksamer machen. Die zentrale Botschaft: Agenten werden die Unternehmensarbeit verändern. Erfolgreich sind jedoch nicht diejenigen, die möglichst viele Agenten einsetzen, sondern diejenigen, die sie an den richtigen Stellen gezielt skalieren.

#359: When was the last time you sat through a 30-minute product demo and walked away actually knowing anything? You would learn more from five minutes hands-on than an hour of watching someone else drive. Now you have help. An agent can watch the 30-minute video, play in the sandbox, read every page of the docs, and come back before you finish your coffee with a verdict - tried it, does not work, next. The agent is the new tire kicker. So if you are a vendor, an open source maintainer, or the person building the internal app nobody outside the building ever sees, the demo you have been giving is aimed at a buyer who already left the room. Your job now is to make life easier for agents. An MCP server, a CLI, skills, an AGENTS.md file, not blocking your own site with Cloudflare when someone's agent tries to read your pricing. Everything that makes a product easy for an agent would have made it easier for a human all along. We just never bothered, because we had months to burn. Now the clock runs in minutes and every corner we cut is suddenly on fire. Three kinds of demo, three different answers. The vendor sales demo is off-putting before it starts - if a website says book a call to try it, Viktor is already gone. Open source barely needs a demo at all: a good README, a quick start, an AGENTS.md, and the agent assembles a demo tailored to your stack, your database, your questions, instead of some generic happy path. Internal is where it gets good, and it might be the one that matters most since exactly zero apps ship without customization. Viktor's bar: stop showing me plans, show me the thing running. Sit the stakeholder down and build it live while you talk. Three days to a prototype instead of 300 pages of PRD. Sandboxes first, demos second - if you cannot spin up a sandbox, you did not build it right. And demo the failure modes, not the happy path, because resiliency is the real selling point now. Disks still fill up. No amount of AI magic empties them for you.   YouTube channel: https://youtube.com/devopsparadox   Review the podcast on Apple Podcasts: https://www.devopsparadox.com/review-podcast/   Slack: https://www.devopsparadox.com/slack/   Connect with us at: https://www.devopsparadox.com/contact/

Joel Brewer, founder and CEO of Brewer Digital, joins Jeff Mains on Future Proof Founder to cut through the AI hype and get practical about what it actually means to build software in 2025. Joel brings a developer's perspective to a conversation that's usually dominated by non-technical voices — and the result is a grounded, honest look at how early-stage SaaS founders can work with engineering teams, avoid costly mistakes, and use AI as a tool rather than a crutch.From rescuing failed offshore builds to helping non-technical founders navigate Claude Code, Joel shares what he's learned from 13 years in the startup trenches. The conversation covers sandbox environments, the evolving definition of a senior engineer, the difference between bootstrapped and VC-backed engineering cultures, and why clear goals — not shiny tools — are still the most valuable thing in any founder's arsenal.Key Takeaways3:46 — Joel shares how getting fired launched his 13-year journey building Brewer Digital from freelance Rails work to a full software development agency for early-stage founders.6:59 — Goals before tools. Joel's antidote to AI-driven chaos: go back to basics. Know where you're headed before you pick up a tool. AI accelerates whatever you point it at — including anxiety.11:36 — The sandbox lesson. Real-world story: a client went rogue with Claude Code and made changes directly to a production database, thinking they were just experimenting. Sandboxes exist for a reason.14:15 — The new definition of a senior engineer. From Google I/O: senior engineers used to be defined by what they could build that others couldn't. Now, they're defined by what they understand that others don't — especially critical as AI writes more code.15:17 — The construction metaphor. AI doesn't eliminate the need for engineers — it expands what's possible to build. We went from huts to skyscrapers with better tools. The same shift is happening in software.19:36 — Communication breaks before code does. When Joel steps into a failed project, the first red flag he looks for isn't the codebase — it's whether communication between founders and engineers has broken down.21:08 — Optimize for the right thing. Feedback loops are powerful — but only if you've already defined what you're optimizing for. Shipping features faster means nothing if the features aren't moving the needle.24:21 — The Southwest Airlines principle. Joel shares a CEO quote about the real goal of flying a plane — not the technical operation, but getting someone home for Christmas. The same applies to product teams: the feature is never the goal.25:52 — Bootstrapped vs. VC-backed engineering cultures. Bootstrappers tend toward capital efficiency and caution. VC-backed companies optimize for velocity and growth at scale. Different goals, different cultures — both valid depending on the mission.30:21 — What gets lost when AI replaces people. When you replace humans with agents, you lose humanity and irreplaceable individual creativity. Joel argues that sitting with a notebook for 15 minutes can still outperform an hour with AI for generating genuinely inspired thinking.33:36 — Advice for non-technical founders hiring developers. Know your goals for the product before the first engineering conversation. Engineers love solving problems — if you don't guide them with clear goals, they'll solve the wrong one brilliantly.35:03 — The future of founder-engineering relationships. AI removes code-writing as the bottleneck. The new bottleneck is communication — rapid testing, alignment, and iteration between product, founder, and engineering.38:24 — Complexity kills growth. AI tools are excellent at adding complexity. Founders who learn to fight that tendency — preferring simplicity and brevity — will have a structural advantage.Tweetable Quotes"AI becomes a tool when you know where you're headed. If you don't know your destination, AI just helps you get lost faster." — Joel Brewer"A senior engineer used to be someone who could build what others couldn't. Now, they're someone who understands what others don't." — Joel Brewer"The sandbox isn't optional. You can ask Claude to change your production database, and it will. Enthusiastically." — Joel Brewer"Complexity kills growth. AI is really good at adding complexity. That's a problem worth solving on purpose." — Joel Brewer"I sat down with a notebook for 15 minutes and had more genuinely inspired ideas than an hour with AI. People can still think. That's not nothing." — Joel Brewer"If your imagination is unlimited, AI doesn't replace you — it just lets you build the skyscraper instead of the hut." — Joel Brewer"We're not very good at predicting the future. And when there's a lot of money involved in selling a specific version of it, that's worth keeping in mind." — Joel BrewerSaaS Leadership Lessons1. Start with goals, not tools. The most expensive AI mistakes happen when founders adopt a tool and then search for a use case. Reverse the order: define what you're trying to accomplish, then evaluate whether AI actually serves that goal. Tool-first thinking creates fast, sophisticated chaos.2. Build a sandbox culture — literally and figuratively. Non-technical founders experimenting with AI-assisted development tools need guardrails. In software, that means sandboxed environments separated from production. In leadership, it means creating space for experimentation that doesn't put the business at risk. Permission to explore is only valuable when the blast radius is contained.3. The new competitive edge is understanding, not output. As AI commoditizes the ability to generate code, copy, and content, the strategic advantage shifts to comprehension — people who understand why something works, what the architecture actually does, and when not to build something at all. Leaders who can explain the machine are worth more than leaders who can only operate it.4. Communication is the bottleneck AI just uncovered. Before AI, writing code was slow enough that communication gaps could hide. Now that code can be written in hours, the limiting factor is alignment — between founders and engineers, product and market, vision and execution. Founders who invest in communication infrastructure will outpace those who just invest in faster tools.5. Tie every feature to a business outcome — before the first line of code. Engineers are problem-solvers. They will build what you put in front of them. If the problem you hand them isn't connected to a business goal, they'll build the wrong thing exceptionally well. Make the goal explicit, measurable, and understood by the whole team before any work begins.6. The human element isn't a liability — protect it. When every company is using similar AI stacks, the differentiator becomes the people. Human creativity — particularly the kind that emerges from unstructured thinking time — is increasingly rare and undervalued. The best leaders will use AI to create more space for deep thinking, not less.Guest Resourcesjoel@brewerdigital.com www.brewerdigital.com https://www.linkedin.com/in/joel-brewer-69680037/ Episode SponsorThe Futureproof Series - https://www.youtube.com/playlist?list=PLfkXKUPZ5xuOqMPR7_gzGybncTtavyR1NThe Captain's KeysSmall Fish, Big Pond – https://smallfishbigpond.com/ Use the promo code ‘SaaSFuel'Champion Leadership Group – https://championleadership.com/https://jeffmains.com/books/SaaS Fuel ResourcesWebsite - https://championleadership.com/Jeff Mains on LinkedIn - https://www.linkedin.com/in/jeffkmains/Twitter - https://twitter.com/jeffkmainsFacebook - https://www.facebook.com/thesaasguy/Instagram - https://instagram.com/jeffkmains

Welcome to Audio EXP episode 341. This week, we explore a digital preservation milestone as the June 1977 debut issue of White Dwarf magazine is added to the Internet Archive's historical library. We also examine the launch of the ENNIEs Emporium 2026 on BackerKit, showcasing how independent publishers are combining forces for rapid retail shipping, and catch up with exclusive interviews featuring the creators of Land of Eem and Courtesans' Guild. Plus, we cover Bronwen's updates on House of the Dragon and Hot Spot. About Audio EXP Audio EXP is Geek Native's podcast. Each week, there's some favourite or exciting geeky news, conventions, interviews, and thought pieces. The average length of the podcast is around 10 minutes. You will find a transcript of this week's podcast and links to the stories mentioned here: https://www.geeknative.com/239618/audio-exp-341-archive-access-co-op-marketplaces-and-award-winning-sandboxes/

We've been running a bit of an Agent Cloud series surveying all the top inference/compute/cloud providers, from Databricks to Daytona to Railway and, even further back, E2B, but we're excited to conclude this series returning to Modal, which has just raised a monster $355M Series C.The cloud was built for developers. But agents are now changing that.The old infra stack was designed for a human who could read docs, reason through YAML, and understand dashboards to figure out what they need when something broke. While this was painful for developers, it worked since they could fill in missing context in their heads.However, agents don't have that luxury. Now in this new era of agents, everything has to be tighter.They need a place to write code, run it, inspect the output, change the environment, debug failures, and try again. Fast iteration and feedback loops with all the necessary context are crucial for agents to operate properly. Furthermore, sandboxes are a clear representation of this shift as agents can easily spin up isolated environments. This programmatic infra even extends to research:Two years ago, we were one of the first to cover Modal with CEO Erik Bernhardsson and Alessio designed our favorite LS thumbnail of all time:At the time, Modal was just a teeny little company with a $17M Series A.Today, fresh off their $355M Series C, Modal is one of the clearest examples of the agent cloud future being built in real time: a cloud platform moving past traditional web app assumptions toward the workloads AI actually creates such as elastic inference, sandboxes, GPU burst, post-training, background agents, and infrastructure that agents themselves can operate.In this episode, Modal CTO Akshat Bubna joins swyx and Vibhu to unpack why AI applications don't fit traditional cloud assumptions, why Kubernetes was never designed for bursty compute-heavy workloads, and why Modal is now shifting from developer experience to agent experience.We go deep on Modal's AI infra stack: serverless functions, decorator-based infrastructure, elastic inference for custom models, GPU snapshotting, DeFlash, speculative decoding, Auto Endpoints, sandboxes, persistent storage, networked containers, private IPv6, RDMA, multi-node training, and Modal's capacity pool across 17 cloud providers. Akshat also explains why RL rollouts can require 100,000 sandboxes, why production agents need hard guardrails, why observability may matter more than reading code, and why AI has made infrastructure exciting again.We discuss:* Why Kubernetes wasn't built for bursty AI workloads* How Modal started as a better runtime before becoming an AI cloud* Why Modal added GPUs before ChatGPT* The shift from developer experience to agent experience* Why observability matters when agents are writing the code* Elastic inference for custom models across audio, video, robotics, and comp bio* GPU snapshotting, cold starts, and why inference workloads are so bursty* Why RL rollouts can require 100,000 sandboxes* DeFlash, speculative decoding, and frontier-level inference performance* Auto Endpoints and making optimized inference easier to deploy* What Modal adds beyond vLLM, SGLang, and raw GPU rental* Modal's 17-cloud capacity pool and supercloud strategy* Networked sandboxes, sidecars, private IPv6, and RDMA* Serverless multi-node training for post-training and research workloads* Auto-research, model-guided sweeps, and agents launching GPU experiments* Compute strategy, capacity planning, and batch tiers* Why production agents need specialized sandboxes and hard guardrails* Modal's take on managed agents, CI, Gitpod/Ona, Python, TypeScript, and Modal BenchAkshat Bubna* LinkedIn: https://www.linkedin.com/in/akshat-bubna-188885103* X: https://x.com/akshat_bModal* Website: https://modal.comTimestamps00:00:00 Introduction00:00:39 Modal's origin and why Kubernetes wasn't enough00:04:32 Developer Experience → Agent Experience00:06:21 Modal's AI cloud primitives00:09:14 Sandboxes, agent loops, and proto-Cognition00:12:12 Elastic inference, GPU snapshotting, and 100,000 sandboxes00:15:24 DeFlash, speculative decoding, and Auto Endpoints00:19:59 Production-grade inference beyond raw GPUs00:22:00 Background agents, Ramp Inspect, and the agent lifecycle00:24:08 Modal's 17-cloud supercloud strategy00:26:40 Networked sandboxes, private IPv6, and RDMA00:32:48 Multi-node training, post-training, and auto research00:37:36 Compute strategy, capacity planning, and batch tiers00:40:55 Open models, real-time AI, and production agent infra00:43:06 Hard guardrails, managed agents, and specialized sandboxes00:46:06 Why AI made infrastructure exciting again00:48:30 Model APIs, differentiated products, and agentic video00:51:50 CI, coding-agent infra, SDKs, and Modal Bench00:57:28 Closing ThoughtsTranscriptIntroduction: Modal, Series C, and the Art PartySwyx [00:00:00]: We're here with Akshat, CTO of Modal, together with Vibhu. Congrats on your Series C.Akshat [00:00:10]: Thank you.Swyx [00:00:11]: Your party yesterday was amazing.Akshat [00:00:15]: Yeah.Swyx [00:00:15]: From all the photos and all the swag.Akshat [00:00:17]: We had a bunch of art installations, which was fun, seeing, like, our products on pedestals next to, like, Rodin.Swyx [00:00:25]: Very nice. Very nice. When you started, it was not the GPU inference company. Maybe it was in your mind. Take us back to the origin story.Modal's Origin: A New Runtime Beyond KubernetesAkshat [00:00:39]: I first met Eric, who's the CEO, through an investor. Back then Eric was already thinking about building, a new runtime, and he got there thinking through why are workflow orchestration products so hard to use. It's because you have to run them on Kubernetes. Kubernetes is hard to manage. It's not built for burstiness and, custom images,Swyx [00:01:03]: YeahAkshat [00:01:03]: It has a terrible developer experience.Swyx [00:01:05]: And I'll, I'll interjectAkshat [00:01:06]: YeahSwyx [00:01:07]: For listeners, who are new, we interviewed Eric two years ago, and there's a bit more of the story there from Spotify and all those things.Swyx [00:01:14]: And I came across Eric through Data Council because he did that talk on the serverless container stack that you guys did, which was like, that was my first like, “Okay, I need to take Modal very seriously” moment.Akshat [00:01:26]: Yeah.Swyx [00:01:26]: But it was still very unclear, like, do I need all this for just my data pipelines?Akshat [00:01:33]: Yeah. initially what we were thinking about was if we build a better runtime, it's a very useful primitive in itself. It's There's a lot of things that, get solved by serverless functions, like you can do, ETL stuff, you can do job queues, you can do all this, like, bursty processing, which it turns out every company had needs for. but then we also were thinking about this as like, this is a primitive that we can build a whole collection of products on, which are very verticalized. So perhaps data engineering would've been the first one, but we were thinking about inference. Back then it was more classical inference, like computer vision stuff and running XGBoosts and whatnot. But we added GPUs to the product a year before ChatGPT came out.From Serverless Containers to GPU WorkloadsSwyx [00:02:19]: Nice.Akshat [00:02:19]: We just didn't think it would be that big of a deal.Swyx [00:02:22]: Yeah, just like add A100.Vibhu [00:02:23]: Was there any, like, early key problem that really sparked off why you built it?Akshat [00:02:28]: Yeah. Primarily it's just, none of the tooling that was out there was built for, one, a really great developer experience, and also there's a general trend of, a lot of the workloads that we were seeing were very. I wish there was a better word for it, but compute-heavy. Like, they need, one, like, need a lot more resources, so you need to burst up and down a lot, versus like Kubernetes designed for, like, slow scaling and, more for, like, web server use cases. And also there's just a lot more specialization in, like, what kinds of environments these workloads run in. Like, we had sometimes they need accelerators, sometimes they need different kinds of images, and this is just like a consistent thing that we saw across a lot of companies. That would be the next step.Software-Defined Infrastructure and Decorator-Based DXSwyx [00:03:13]: Yeah. Yeah. Be nice. I don't know how much this factored into the early story, but I wrote a post when I was at Temporal about infrastructure, software-defined infrastructure or something like that.Akshat [00:03:22]: Yeah, the self-provisioningSwyx [00:03:23]: Self-provisioning.Akshat [00:03:24]: Yeah.Swyx [00:03:24]: Yeah. I can't even remember my own post.Swyx [00:03:26]: And then you put me on the landing page.Akshat [00:03:28]: Yeah. We really like, the term and so we stole it.Swyx [00:03:32]: Because you had the insight that everything can just be in decorators co-located with the code, right?Akshat [00:03:37]: Yeah.Swyx [00:03:37]: Was that a big part of the originalAkshat [00:03:39]: YesSwyx [00:03:39]: Story or it was just like a DX layer?Akshat [00:03:41]: That was, really important because we really didn't want people to spend, so much time, writing YAML, and it seemed like you could really condense the surface area of what you're doing, put it in code so you can operate on it just like you operate on other code, and like build stuff that's more expressive and dynamic. and so yeah, that was always a very important part.Swyx [00:04:04]: Then the pushback is this is a DSL.Akshat [00:04:07]: Yeah.Swyx [00:04:07]: It's you're closed source. I am locked into Modal.Akshat [00:04:11]: Yeah. We never really got pushback for that because the nice thing about Modal is you can bring whatever code you have, and sure, the DSL is at the configuration layer for, what hardware you're using, how you're scaling things up, but you still own the code.Akshat [00:04:27]: And that's, that's been an important, part of our story, even as we do inference now.Swyx [00:04:32]: Yeah.Vibhu [00:04:32]: How much of do you think still stays the same today? Like if you were to build something today, DevX very important, but I feel like, a lot of this has been changed with just hook it up to an agent, have Claude Code, have Codex implement a tool. there's very agent native primitives that are different than if I'm doing this myself, right?Developer Experience → Agent ExperienceAkshat [00:04:54]: We've changed our SDK team to think about agent experience instead of, developer experience and we think that the same benefits that apply for DX also apply for AX, which is why would you have an agent read through hundreds of Kubernetes files and like write YAML that's not even typed when it can make a couple of changes in a decorator and it gets this self-provisioning runtime of, being able to see its changes live in action? yeah, it just seems from the customers we talk to, they find Modal is much faster for agents to use versus operating on a different substrate.Swyx [00:05:34]: Yeah, because like you, again, you co-locate the infrastructure requirements to the code that runs it.Akshat [00:05:38]: Yeah.Swyx [00:05:38]: Well, the negative thesis now is that nobody's looking at their code anymore, so there's no point.Akshat [00:05:44]: Yeah, people aren't looking at code. one thing we still see is really important is observability.Swyx [00:05:51]: Yeah.Akshat [00:05:51]: Like how good is your dashboard? And of course, like we have, we push a lot of it to the CLI so the agents can do their own investigation, but you still need humans to go interpret what's going on and, make judgment calls and whatnot. and that's I feel like, Maybe more important now than looking at the code itself.Swyx [00:06:11]: Yes, because like, you can try to treat the code as a black box and then use, see the observable action that comes out of it, and then just prompt a change.What Modal Is For: AI Cloud PrimitivesAkshat [00:06:21]: Yeah.Swyx [00:06:22]: So I think it takes a bit of restraint to not specialize, to say, “I want to ship a new primitive,” and then just be general purpose.Swyx [00:06:31]: People ask you, “What are you for?” You're like, “ I don't know. We can do this, we can do that.”Vibhu [00:06:36]: Well, I'd be curious to see, like, okay, if we were to ask you, like, what is Modal for even at a high level? There's a lot you guys do, sandboxes, GPUs, everything. How do you answer?Akshat [00:06:46]: Modal is a cloud platform that's built for, where we've built the primitives from scratch for AI applications. and right now it covers, inference, training, batch processing, and sandbox workloads.Akshat [00:07:00]: But we're building a lot moreSwyx [00:07:02]: I noticed you didn't say web server, so there is still a role for, like, the always-on large-scale Kubernetes type things.Akshat [00:07:09]: Yeah, absolutely. We're, we're not trying to compete with the renders of the world, because yeah, we think the differentiator for us is the, are the workloads that need specialized compute, need to scale up and down a lot. yeah, they're, they're, they're just shaped differently.Working Alongside Frontier StartupsVibhu [00:07:26]: I think you're building a lot of it alongside the startups, right? They're innovating quite a bit, even in your, like, latest blog post. Like, even in the series C, the customers that you mention here, the cognitions, technical ones, ramps and whatnot, they're, they're innovating with you, right? And that's not something AWS is doing directly with.Akshat [00:07:45]: Yeah, absolutely. I think, this is again classic. We're a small team. We can move really fast. our engineers are working with our customers and figuring it out. Yeah.Swyx [00:07:54]: So my first week at Cognition, I walked in, there was someone wearing a Modal shirt. I was like, “What are you doing here?” They're like, “Yeah, I just. I am embedded inside of Cog.”Akshat [00:08:05]: Yeah, I think that was Peyton. We sent him overSwyx [00:08:07]: Yeah.Akshat [00:08:07]: Because, the latency of communication was too high otherwise.Swyx [00:08:12]: Yeah, distributed node, you have to - you have to place one and collocate.Vibhu [00:08:16]: Yeah.Swyx [00:08:16]: So I had a, I had direct personal experience, right? So I worked on smol developer three years ago. it was inspired by Claude 1. I think you onboarded me at some point, like, just before, and I was like, “Oh, like, I need some bursty compute. Like, I was just gonna try using Modal.” And it was a, it was a pretty pleasant experience. apparently, I showed up in the board meeting, like the analytics.smol developer, Sandboxes, and Proto-CognitionAkshat [00:08:39]: Yeah, you blew up on Hacker News and,Swyx [00:08:41]: YeahAkshat [00:08:41]: We got a big traffic spike. I. I think the way you used smol developer was Modal functions for running stuff, which was. Like, the, that was a good use case. but then, yeah.Swyx [00:08:53]: Yeah. That - So to me, that was proto-cognition.Akshat [00:08:55]: Right.Swyx [00:08:56]: If only I had, like, stuck to it.Swyx [00:08:58]: Like, that was like, if - did you say draw the tech treeAkshat [00:09:00]: AbsolutelySwyx [00:09:00]: You're just like, “Yeah, like, probably this will happen.”Akshat [00:09:02]: Yeah. Like, he was so close. You were just rebuilding upon usSwyx [00:09:04]: I just didn't realize.Akshat [00:09:05]: But the funny story there is at the same time, we were talking to a bunch of customers who needed something like sandboxing.Swyx [00:09:14]: Yeah.Akshat [00:09:14]: This is like twenty-three.Swyx [00:09:15]: Yeah.Akshat [00:09:16]: So we builtSwyx [00:09:17]: You introduced a new API right after that.Akshat [00:09:18]: Yeah.Swyx [00:09:19]: Yes.Akshat [00:09:19]: Like, we built sandboxes in May of twenty-three before anyone was even knew this was gonna be a thing. And the first example we published was, we took smol developerSwyx [00:09:28]: Smol developerAkshat [00:09:28]: And put it in a loop, so the agent can iterate on itself.Swyx [00:09:33]: Loops are hot these days.Vibhu [00:09:34]: It's the looper.Akshat [00:09:34]: Yeah.Vibhu [00:09:35]: Loops in. When was this, twenty-three?Akshat [00:09:38]: Yeah.Vibhu [00:09:39]: A small check.Akshat [00:09:39]: Yeah.Swyx [00:09:39]: It's like twenty-three. so the. the, those for listeners, like, the problem was the models are not built for any of this, right?Swyx [00:09:46]: Like, you're just trying to like. They're not post-training to understand, like, looping and, like, self-correction and tool calling was there, but, like, also not that great.Akshat [00:09:55]: Yeah.Akshat [00:09:55]: I don't remember if you used tool calling in this one, but yeah, the models would just diverge after like ten iterations and not produce anything meaningful.Swyx [00:10:03]: Yeah. But like, then. So okay, like now talking to myself three years ago, the answerVibhu [00:10:08]: Of course they will get betterSwyx [00:10:09]: Collect all the failures, build benchmark, and then collect all the, examples, build the RL environmentAkshat [00:10:15]: RightSwyx [00:10:15]: Sell it for like ten billion dollars to Meta.Swyx [00:10:17]: And then also train a model and then sell that for sixty billion dollars to Elon. And this isAkshat [00:10:23]: Yeah, of courseSwyx [00:10:23]: The funny machine. Like, it's like, it's about the hardware.Akshat [00:10:28]: It's hard to have that inherent conviction that the stuff will get that much better.Swyx [00:10:33]: In retrospect, it's so f*****g obvious.Akshat [00:10:36]: Fair enough.Swyx [00:10:37]: Like, what else were we doing back then? I don't know. anyway. Yeah. So this. That was the start of your sandboxing journey, right? I feel like it didn't blow up until, like, last year.Akshat [00:10:49]: Yeah.Swyx [00:10:50]: So there was like a couple years of quietness.Akshat [00:10:52]: Exactly, yeah. We wereVibhu [00:10:53]: I think very underrated product value. Like, my experience with Modal, Charles, before he had joined Modal, met this guy at a hackathon, and he really insisted we wanted to run some small model, not hosted anywhere, and he's like, “ there's this cool company, Modal. They'll like spin up a GPU sandbox, we can throw it on there. They'll take a Hugging Face link.” And like there's so much value just right there, right? Like instant hosting, spin it up, spin it down. It'll stay cold, but we run the demo a few days later, it'll come back up and like all this stuff in retrospect, like it's still what we needed like today.Akshat [00:11:27]: Yeah, it's still needed today. workload shapes have changed a lot as, we run stuff for people with really massive production scale and, there it's it's not about scaling from zero to one, but it's how do we scale really elastically, from like thousand to fifteen hundred GPUs very quickly in a given region. It's the same shape problem.Elastic Inference, GPU Autoscaling, and Custom ModelsVibhu [00:11:50]: Okay. So you look at, say, Cursor Composer, right?Akshat [00:11:53]: Yeah.Vibhu [00:11:53]: They had a. “We'll do RL on a model every couple hours.” you guys have a whole version of RL inference gym and whatnot.Vibhu [00:12:01]: When you look at workloads like that, you're doing train runs where you need to scale up, scale down every hour thousands of GPUs, right? That's the example for we do need it, right?Akshat [00:12:12]: Yeah. Well, so I'll, I'll take a step back and, maybe talk about like how people use Modal today. because our biggest use case is, elastic inference. And the thing we first found product market fit, with was inference for custom models. So we stayed away from the LLM space, and we were serving companies like Suno for audio, Runway for video, robotics, comp bio companies that train their own model elsewhere. But Modal is the best black box that for deployment, scaling to however many GPUs you need as your traffic pattern changes. And we saw all of them like have a very unpredict- predict- predictable, traffic pattern. it's like diurnal. It's Some days, like the company will do a launch and, they'll need like, way more. And it's not just one model that they deploy. They-- all these companies deploy, lots of different models in different regions, and so the autoscaling problem becomes even harder because then you have to scale within a certain region, and those cycles are offset. So different times you scale up in different regions.Akshat [00:13:20]: So that's like our sortVibhu [00:13:22]: And thatAkshat [00:13:22]: YeahVibhu [00:13:22]: That in and of itself is a huge category. There's a bunch of inference providers which, provide this fireworks, does this as a service together, whatnot, Base10. that's carved into its own niche for language models, at least right now.Akshat [00:13:36]: Yeah. the thing that we have specialized in is the autoscaling aspect.Vibhu [00:13:41]: Yeah.Akshat [00:13:41]: Because we found that it's not universally true that everyone else can autoscale, and we've gone deeper into it on the tech side by, we've incorporated GPU snapshotting into the product so we can take the GPU state, like your torch.compile model, snapshot it, and the next cold start is way faster. And so going back to your question, it's That's why you need a lot of burstiness for inference. But then people also do a lot of demand training, like for RL stuff, your rollouts are bursty, as you said. People also do a lot of batch jobs. So we'll see, a lot of companies, before they have a training run, they'll need thousands of GPUs to run encoding or something like that. And I think those things are much more bursty than. I agree that agents are not that bursty. sandboxes are, except when you're doing RL. RL is justRL, Batch Jobs, and 100,000 SandboxesVibhu [00:14:28]: Or commerceAkshat [00:14:28]: Insanely bursty.Vibhu [00:14:29]: Yeah.Akshat [00:14:30]: Yeah. Like when you're doing, rollouts, you sometimes need a hundred thousand sandboxes in your sandboxes.Vibhu [00:14:37]: Yeah. I'm curious if you've seen early sparks of continual learning. There are some people, like our friends, ngram, recently announced thisAkshat [00:14:45]: YeahVibhu [00:14:45]: They're, they're trying to do training. That also seems like a different workload, right? If you're doing training twenty-four/seven per se, there's a very weird dynamic of how you're using GPUs between people and whatnot, but seems like something you guys would work for.Akshat [00:15:00]: As you said, we're, we're fortunate to work with a number of, customers at the frontier and grab some of our customers. and they are taking the primitives we have, and trying to use them in very interesting ways, like continual learning. It's possible as the stuff gets better, some of that will be part of, our offering as well if, more people need it. but we're, we're just waiting to seeVibhu [00:15:23]: YeahAkshat [00:15:23]: How it shakes out.Vibhu [00:15:24]: Is there a primitive that you added after sandboxing that was the next step in the story?LLM Inference, DeFlash, and Speculative DecodingAkshat [00:15:32]: I guess we've been going much deeper into LLM inferenceVibhu [00:15:35]: YeahAkshat [00:15:35]: Because we realized that some of the advantages we have with like autoscaling, again, especially in different regions and whatnot, are, not present elsewhere. and the place where we had a gap was we weren't, working on the model layer itself. Like we were a black box. And, we realized that, we can get to frontier-level model performance, with, by having great people who work on this. And, we've been open sourcing a lot of our work, in terms of, Recently, we, shared our work on DeFlash, which is a block-based, speculator, and we've open sourced, all of it. So, you can - By using open source DeFlash, you can get the same performance as you would with one of the proprietary providers. And the next thing we're thinking about hereVibhu [00:16:23]: I thought this wasAkshat [00:16:24]: YeahVibhu [00:16:24]: An interesting blog post as well, right? Like, I think in here you make a claim that. Not a claim, just that how effective speculative deco-decoding really just get to.Akshat [00:16:33]: Yeah.Vibhu [00:16:33]: Anything you wanna point out from this around, what people should know?Akshat [00:16:39]: Yeah, absolutely. the high-level summary is, it would help to describe what speculative decoding is.Vibhu [00:16:44]: Yes.Akshat [00:16:44]: I will, yes.Vibhu [00:16:45]: I think, likeAkshat [00:16:46]: YeahVibhu [00:16:46]: So we've covered like Eagle and all thisAkshat [00:16:47]: YeahVibhu [00:16:47]: Like Hydra and all those things, but it was like two years ago.Akshat [00:16:51]: Yeah.Vibhu [00:16:51]: I think it doesn't hurt, right?Akshat [00:16:52]: Yeah. Speculative decoding is you have a smaller model, called a draft model, predict tokens ahead of the bigger model, and then you have the bigger model, verify all of this, all the tokens are predicted. And the reason it's faster is if you're predicting, one token at once, you're bound by memory bandwidth. But if you can batch the verification of, the draft model, then you're much more efficient using compute, and it's faster, and as long as your draft model is producing a lot of tokens that can get accepted, which is called the accept length, you can get a speed up that's, multiple times of, the original model speed. and well, that's what we highlight here. It's Like people talk a lot about we made these kernels faster and whatnot, but improving kernel will only give you like few percentage points of improvement, and, increasing accept length, literally is a multiplicative decreaseVibhu [00:17:47]: Like two to four X.Akshat [00:17:48]: Yeah, exactly.Vibhu [00:17:48]: Without much head-on performance.Akshat [00:17:50]: Yeah. I think it may - you are running a second model, right? So it may be something more expensive in the compute,Vibhu [00:17:57]: I meant quality performanceAkshat [00:17:58]: Probably not by muchVibhu [00:17:58]: But yeah. I thinkAkshat [00:17:59]: So there's no drop in quality performanceVibhu [00:18:01]: YeahAkshat [00:18:01]: Because you're always. You're never accepting a token that the big modelVibhu [00:18:04]: It's strictly betterAkshat [00:18:05]: YeahVibhu [00:18:05]: Or it's same.Akshat [00:18:06]: Exactly.Vibhu [00:18:07]: Right. Yeah.Akshat [00:18:08]: And so we've been working a bunch on DeFlash, which is a block-based speculator. so it's instead of predicting, one token at a time, it's predicting a block. And we've been open sourcing our work with it. The next thing for us here is for helping people train speculators and custom models. it's it's something that traditionally is very forward-deployed engineering driven, support deployed, engineer driven, like you work with customers and help them do that. And our vision for. This is why we launched Auto Endpoints, is we want to make frontier-level performance available to everyone. And so, we mentioned this in the announcement, we teased it. The next thing we're, we're launching is, as you run an auto endpoint, we shadow trafficAuto Endpoints and Frontier-Level PerformanceVibhu [00:18:54]: Do you want to explain what auto endpoints are?Akshat [00:18:57]: Yeah.Vibhu [00:18:57]: I lovely, yeah.Akshat [00:18:58]: Yeah. So, this is, I guess, going back to your Modal is you touch the code, but, sometimes people don't wanna touch the code, and they wanna get started with an endpoint that works and has all the great performance and, scalability that Modal has. So we've made that easier with, a way to create an endpoint from our UI, from the CLI, that has all of our optimizations that we talked about, like the DeFlash stuff already baked in, and there's full transparency. So we give you the code, you can go run it yourself, and if you want, you can eject out into the full Modal experience, which we see as people get sophisticated, they do wanna tweak the models, they wanna, fine-tune stuff. You can still do all of that. It's it's not a black box. And yeah, the next thing, as we teased later in the post, is how do we give you value even beyond this in terms of having your draft models evolve as your data distribution evolves, again, without having to talk to a person and, yeah.Vibhu [00:19:59]: I guess just to understand it directly, you have the GPUs, you have an endpoint that's compatible, you serve open model. If someone was to do this themselves, what's the delta that you guys provide? So you do a lot of open source great work on effective inference. how does it compare to, say, I take the same model, 5.2 FP8, take shelf inference engine, vLLM, SGLang, get compute of similar capacity, similar cost. What's the delta that plugging into something this, like this offers outside of the benefit of, scaling?Production Inference Beyond Raw GPUsAkshat [00:20:34]: It's interesting because we've taken the approach of open sourcing our contributions and upstreaming them. we work closely with the SGLang team. We want the improvements that our team, comes up with to be, there in open source for others to use, even outside of Modal. The benefit to us is we have a team that has significant expertise in terms of if you do have something that is not there, our team can help you get that performance, first. the other thing is with these endpoints, we are way more elastic, as you said, than, anyone else, and you have true scaling to zero. you have true, burstiness, and in practice, that matters a lot more to people than just finding, the GPU and, running Modal code on something.Vibhu [00:21:20]: Yeah. And I will say it's not that straightforward to just. like what I said is easier said than done, right?Akshat [00:21:26]: Yeah.Vibhu [00:21:27]: It's I think still for the average person, still hard to just gut check using different. There's, there's quite a bit of combinations you can make there. the trade-offs aren't really known at face value.Akshat [00:21:40]: Yeah. it's it's not just that. I think it's it's that running production-grade inference is a hard infer problem.Vibhu [00:21:49]: YeahAkshat [00:21:49]: Even if you subtract out the autoscalingVibhu [00:21:50]: YeahAkshat [00:21:51]: Is controlling things like tail latency and, making sure every, request is delivered at least once and whatnot.The Model and Agent LifecycleVibhu [00:22:00]: There's a lot of innovation that you can do here. I think, it's very interesting that you're starting to encroach on, like as you become a full cloud, you're starting to encroach on other people's turf.Vibhu [00:22:09]: What will you not do?Akshat [00:22:13]: Well, we wanna follow our users and, make sure they get like a platform that has everything that works well together. so right now we're focused on the model lifecycle and the agent, lifecycle. so both like going from data prep to training to inference, and then also if I want to deploy a background agent, let's say, sandbox, do persistent storage, a whole bunch of other stuff.Vibhu [00:22:38]: We talked to Cole, who did, OpenInspect. Yeah.Akshat [00:22:42]: Yeah.Vibhu [00:22:42]: And RealInspect also is on Modal.Akshat [00:22:44]: Yeah. So Ramp Inspect was a great example of a background agent that was really successful because they, were able to use some of the primitives like snapshotting and fast scaling to just have something that feels really reactive and works well.Ramp Inspect and Background AgentsVibhu [00:23:02]: Yeah. That's the new CTO of, Ramp right there.Akshat [00:23:05]: Yeah, Rahul.Vibhu [00:23:08]: It was really fun. yeah, okay, I think, all very bullish. Like, one of my reflections was also I did not originally. So when I met you guysThe Inference Inflection: CPU, GPU, and Co-LocationVibhu [00:23:19]: You weren't that much in the GPU game, and now you're all about, inference. And one of the points that I hinged on for Jensen's keynote at GTC this year was, what we're calling like the inference inflection, right? That let's say in AI workloads or machine learning workloads, it used to be like, let's call it eight to one GPU to CPU, and now it's more like one to one, which is like a interesting. Like, - because of how much agents are blocked or call out to this, to CPU heavy stuff the actual, like, limiting factor, like, swings back and forth from GPU to CPU a lot more than it used to be all GPU and then occasional CPU.Akshat [00:24:01]: Yeah.Vibhu [00:24:02]: GPU, CPU. And now it's like just constantly, and you just have to locate everything.Seventeen Clouds and the Supercloud StrategyAkshat [00:24:08]: Yeah. And that's one of the things that, again, we see as, something appealing about Modal, which is we've built this capacity pool that spans, 17 cloud providers, so we're, we're very good at Running on various kinds of cloud capacity across the worldSwyx [00:24:24]: You don't have your own data centers?Akshat [00:24:25]: We don't have our own data centers. We just run across a lot of neo cloudsSwyx [00:24:29]: Yeah. AreAkshat [00:24:30]: Metal providers.Swyx [00:24:30]: Yeah. Question mark.Swyx [00:24:31]: Yeah. You're, you're running the math, and you're like, “What's the cutover point where you're like.”Akshat [00:24:36]: Yeah, it's a good question. part of it is we see our differentiator in the software layer, and, being capital light and focusing on the software helps us move really fast. so far it's worked out well because there are so many other people building data centers that we're able to work effectively with them, and again, focus on what makes us, special.Swyx [00:24:55]: Yeah.Swyx [00:24:56]: 17 gets you into, like, the local providers sometimes. LikeAkshat [00:25:00]: The,Swyx [00:25:01]: Which was the most interesting one?Akshat [00:25:02]: There are a lot more neo clouds than you expect, and they all have various degrees of, various levels of reliability. And, that's why it's something we've invested a lot of time in, is building our own reliability layer on top. so if the GPU falls off the bus or something happens, we user workloads are not affected, and that lets us use a lot more capacity than,Swyx [00:25:30]: YeahAkshat [00:25:30]: You as a user would be able to.Swyx [00:25:32]: It's a useful thing to have because like now everyone knows, like, what layer you are and, like, you optimize for being the super cloud of all clouds.Akshat [00:25:41]: Yeah. That's, that's, that's the idea. and so I guess when you mentioned colocation, that's, that's another interesting thing where, one thing we've seen is people come to us when they want, very specifically located, CPUs or GPUs, like they wantSwyx [00:25:57]: Oh, they pin it in likeAkshat [00:25:58]: YeahSwyx [00:25:58]: EU?Akshat [00:25:59]: Exactly. Or EU, US.Swyx [00:26:01]: Right. Data resiliencyAkshat [00:26:02]: AustraliaSwyx [00:26:02]: Locality thing or performance or what?Akshat [00:26:04]: It's either data locality or latency, yeah.Swyx [00:26:07]: Yeah.Akshat [00:26:07]: Like, you want your. They're running sandboxes and model. They want them to be right next to aSwyx [00:26:10]: Yeah, it's easy thenAkshat [00:26:11]: YeahSwyx [00:26:12]: To. That is important in all those things. and so, like, you've accidentally, I don't know if it's accident, but, like, you've built the perfect primitive for agents to express themselves. And then, like, it's almost very funny how every extra development just involves more file system, just involves more CPU.Akshat [00:26:30]: Yeah.Swyx [00:26:31]: Just like the things that you already have. I don't know much about, if there's any, like, networking usages that are interesting, but you've also done some good work on networking.Networking, Sidecars, Private IPv6, and SandboxesAkshat [00:26:40]: Yeah, that's exactly right. Like, we're just taking compute storage and networking and building stuff on that layer, for, again, the stuff people need.Swyx [00:26:49]: YeahAkshat [00:26:50]: We see a few interesting networking things coming up. one is people want networked sandboxes. so we haveSwyx [00:26:57]: For like a Docker cluster type thing.Akshat [00:26:59]: Yeah.Swyx [00:26:59]: Sorry, Docker Swarm. Oh, f**k. What is it called?Akshat [00:27:02]: Compose.Swyx [00:27:03]: Compose type thing.Akshat [00:27:04]: Yeah. So if you want Docker Compose, our sandboxes now support, this thing called sidecars. So you can. A sandbox is a pod of containers, and you can run multiple containers in, a sandbox. also useful because, going back to networking, people want a lot of control over, outbound networking from a sandbox.Swyx [00:27:23]: Yeah.Akshat [00:27:23]: Like, they might wanna run a middle proxy for, like, maybe logging stuff for RL or, controlling how egress can happen to a domain, injecting credentials. and yeah. So we've, we've had to build a lot of that stuff ourselves.Swyx [00:27:38]: Yeah.Akshat [00:27:39]: But then also sometimes people want, sandboxes spanning multiple nodes to talk to each other, which is an emerging thing we're seeing. We have support for that for a different reason, and yeah, we'll see if that becomes stable.Swyx [00:27:52]: Like, just an open socket. It's a. This is directly like mTLS.Akshat [00:27:56]: We do support that, which is you can, expose a tunnel inside a sandbox.Swyx [00:28:01]: Yeah.Akshat [00:28:01]: And then you can either expose it to public internet or it can be, you can add like a HTTP, auth layer above it. But we have this thing called I6PN, which we haven't talked about, which is this, like, overlay network using IPv6 addresses. so if Modal containers, within the same workspace, when this is enabled, can address each other using this private IPv6 address, and no one else can.Akshat [00:28:28]: So it's like private networking, for containers. We built it because we needed it as a primitive for our distributed training product. so we have this other feature, which is you can add a decorator to a function, and you get a cluster of GPUs. and they have RDMA networking. so you can run a distributed training job, that's truly serverless. and we did the overlay network for that. But then we've seen that people are using it for other reasons, and, I'm intrigued to yeah, what would people do with it.Swyx [00:28:59]: Build primitives and let people figure it out, right?Akshat [00:29:01]: Yeah, exactly.Swyx [00:29:02]: You put out a pretty interestingAkshat [00:29:03]: They're like, they read the docs webpage. Let me use thatSwyx [00:29:06]: YeahAkshat [00:29:06]: Something they never intended to work. This is literally not even in our docs page. People somehow found it, and they're using it.RDMA, Memory Movement, and Distributed TrainingSwyx [00:29:12]: Huh.Swyx [00:29:14]: The way you portrayed it with, like, RDMA versus TCP, like, very well laid out, but just the transfer speed change at scale for RL, like yeah, you have it, you have it built in. I'm sure someone found it. It's found it to be a lot more efficient before you made a thing out of it, right?Akshat [00:29:32]: Yeah. And not to split hairs, I guess the overlay network is the TCP overlay network.Akshat [00:29:39]: The reason we have that is you need that to do the key exchange for RDMA before you set up the RDMA network on top of that. but then people found the TCP part.Swyx [00:29:48]: Can I tell you, this is like a big aha moment for me becauseAkshat [00:29:51]: YeahSwyx [00:29:51]: So I review 2,200 submissions for the World's Fair.Akshat [00:29:56]: Yeah.Swyx [00:29:57]: And then I got this from John OsterhoutAkshat [00:29:58]: HuhSwyx [00:29:59]: Who I don't know if. Do John Osterhout by name?Akshat [00:30:01]: The name sounds familiar.Swyx [00:30:02]: He published a. He's a well-known professor, published a lot of interesting software design books, and this is the talk he chose to submit, is on RDMA at Inference. And I'm like, you wouldn't think that this guy, who is like operating systems guy, would care about RDMA.Akshat [00:30:20]: I, it makes sense to me because I,Swyx [00:30:24]: This is the cloud, right? YeahAkshat [00:30:25]: Like, the way you move around your KV cache and how efficiently you can do it, how efficiently you move, your weights from your training GPUs to your inference GPUs in RL is there's a lot of degrees of freedom, and it is a systems problemSwyx [00:30:41]: YeahAkshat [00:30:41]: Moving memory aroundSwyx [00:30:42]: YeahAkshat [00:30:43]: Scheduling.Swyx [00:30:44]: This shows you how primitive my understanding of networking stuff is.Swyx [00:30:46]: Is this like the domain of WireGuard as well?Akshat [00:30:50]: Not quite.Swyx [00:30:51]: It's adjacent?Swyx [00:30:53]: Explain everything.Akshat [00:30:54]: Sure.Swyx [00:30:56]: How do we move memory around GPUs?Akshat [00:30:58]: Well, so sorry. Yeah, that is memory. Sorry, I was talking more, and maybe I was talking like five minutes back, about the private IPv6, addressing that you've set up.Swyx [00:31:09]: Yeah.Akshat [00:31:09]: Is it like it's a VPN?Swyx [00:31:10]: Yeah, it is like a VPN, and yeah, WireGuard is, yeah, you're right. It is,Akshat [00:31:16]: Right. Yeah, you already moved on to new topicsSwyx [00:31:17]: A similarAkshat [00:31:18]: OkaySwyx [00:31:19]: In the same space, WireGuard is, encrypted and this is,Akshat [00:31:23]: And you don't need encryption.Swyx [00:31:23]: Yeah.Akshat [00:31:24]: Yeah.Swyx [00:31:24]: This is not encrypted. that's the main difference. This is TCP and we have eBPF programs that will reject or allow the TCP connection based on whether you're allowed to do it.Akshat [00:31:35]: Used to involve a full sidecar, but now you have eBPF in the Linux kernel.Swyx [00:31:39]: Yeah.Akshat [00:31:40]: Yeah. I don't know if this is a natural follow-on to the topic of like my skepticism on distributed training is that while, like, people spend a lot of money on, like, cables to hook up GPUs, and even that is not, like, fast enough, and that's the bottleneck, is your networking fast enough?Swyx [00:31:59]: Yeah. So I guess you're talking about fully distributed training like, Dialog or something which is like cross data centerAkshat [00:32:06]: That would be, yes.Swyx [00:32:07]: That's the extreme.Akshat [00:32:08]: Yeah.Swyx [00:32:08]: You're in the middle, and then other people would have like the Mellanox cables up in, like, their actual data center.Akshat [00:32:14]: When you run multi-node training on Modal, RDMA, I think Mellanox, is, or InfiniBand is like a, is all seen as RDMA. but it's a way to bypass the TCP networking stack and, transfer, stuff much faster, between one node, to the other. And we have I think like 3 terabit per second, internal networkingSwyx [00:32:40]: OkayAkshat [00:32:40]: Which is the standard that's needed.Swyx [00:32:42]: Okay. So I misunderstood whatAkshat [00:32:43]: 50Swyx [00:32:43]: What part of the stack you wereAkshat [00:32:44]: 50 gigs overSwyx [00:32:45]: YeahAkshat [00:32:45]: If you wentSwyx [00:32:45]: YeahAkshat [00:32:46]: RDMA.Swyx [00:32:46]: Okay.Swyx [00:32:48]: Yeah. I, very impressive work.Multi-Node Training, Post-Training, and Auto ResearchSwyx [00:32:52]: So effectively you're extending like the model philosophy to the training cluster, like, yeah.Akshat [00:32:59]: Yeah. And we're, we're not going for like large scale training runs. the thing that we've built multi-node training for is, we see a lot of, smaller scale post-training. like, people are post-training like medium sized fund models, so they can, get higher quality on inference. this is a perfect fit, for something like that.Swyx [00:33:21]: Yeah. That is my impression of how a lot of these labs explore branches in post-training and then eventually merge whatever they find in.Akshat [00:33:31]: Yeah. The other use case we've seen for multi-node training is even if you have a big cluster, your researchers are still doing small runsSwyx [00:33:38]: YesAkshat [00:33:39]: Having elasticity thereSwyx [00:33:40]: Right, sureAkshat [00:33:40]: Matters a lot more.Swyx [00:33:41]: Yeah. the, like, this is like the current limiting factor for auto research, which is like you need to give your model some GPUs in order for it to completely run.Akshat [00:33:51]: We have a blog post on auto resource and model is,Swyx [00:33:55]: YeahAkshat [00:33:56]: Yeah, like, turns out to be pretty good substrate for that.Swyx [00:33:59]: So my impression is auto research means many things, likeAkshat [00:34:01]: YeahSwyx [00:34:01]: Anything that Andrej coins. Right now it's still science fair, right? Like not like, I don't know how many people are doing this.Akshat [00:34:08]: We're having a golf.Swyx [00:34:08]: Yeah.Akshat [00:34:09]: I thought the same thing.Swyx [00:34:11]: Yeah, you would know.Akshat [00:34:12]: We, like, our internal both training and inference teams use this the general shape of this quite a bit. like we have this one internal repo called auto inference, which essentially we've automated our own forward-deployed engineering efforts using, this harness, which is, the agent will just spin up a sweep of different things. It'll even run like, NVIDIA inside profiler and it'll like tweak configs and it'll arrive the right thing. it'll change your GPUs both from H200 to B200, and works really well.Swyx [00:34:47]: Nice.Akshat [00:34:47]: So yeah.Swyx [00:34:48]: By the way, I enjoy that your forward-deployed engineering is so technical that you have to do these things.Swyx [00:34:52]: It's very different from forward-deployed engineering from other people.Akshat [00:34:54]: Yeah. For our forward-deployed engineering team is, essentially they're like applied inference researchers or applied training researchers.Swyx [00:35:02]: Someone told me like they have to be able to build, but they also have to be able to sell. do they have to sell or are they like they're good, they're just like post-sale type of thing?Akshat [00:35:09]: It does, being able to talk to a customer and engage effectively with themSwyx [00:35:13]: YeahAkshat [00:35:13]: Matters a lot.Swyx [00:35:14]: They want the same thing.Akshat [00:35:15]: Yeah.Swyx [00:35:15]: ?Akshat [00:35:15]: But it's it's not really a sales, thing. We pair them with-- We have solution architects as well that are more on the sales side.Swyx [00:35:23]: Okay. Let's spend a bit more time on auto research. This is a big focus for for this year. Where does this go? like, have people explored enough? Like, there's all these beautiful charts of like improve and then level off a bit and then you find the next thing. Is this one abstraction up from normal training? Is that how we think about it, or do you think about it differently? Like model level training versus high, like driven hyperparameter search.Auto Inference and Modal BenchAkshat [00:35:51]: Yeah, like,Swyx [00:35:51]: Someone, some people call it like neural architecture search or whatever, right? Like.Akshat [00:35:54]: Yeah, - So the stuff I've seen people do with it is nowhere on the architecture level. It's pretty much tweaking parameters, but it's it's a hyperparameter sweep that's guided by some model intuition, so it's like much more efficient than, whatever other, sweep you would have.Swyx [00:36:12]: Yeah, it's just, it's just a question of where you want to spend your compute?Akshat [00:36:16]: Right.Swyx [00:36:16]: ‘Cause yeah, you can just throw infinite amounts of money on this and somehow you'll bang out Shakespeare?Akshat [00:36:22]: Yeah, infinite monkey.Swyx [00:36:24]: Yeah, so like the very good for model. and I think it's also very important that agents can spin up other agents, can spin up their infrastructure. Like very good for you. how good is our LLMs at generating model code? Like the benefit of existing LLMs is that you are in the data.Akshat [00:36:42]: Yeah. They're, they're surprisingly good. I think like pre Cloud 4 they were not, and then now they're able to shot, stuff out of the box. But we're playing around with releasing like a Modal Bench for like the harderSwyx [00:36:55]: YeahAkshat [00:36:55]: Things, that the LLMs cannot do yet and maybeSwyx [00:36:59]: What's an example of that?Akshat [00:37:01]: I think the things that- Sometimes agents struggle with, without right guidance and a skill is, how to, use the rest of our observability. Like how to. Something is failing, like how do you look at the logs and then update the right thing? It's reasoning about that. But they're able to shot, likeSwyx [00:37:23]: Yeah. You can just add a skill to it?Compute Strategy and Capacity PlanningAkshat [00:37:26]: Yeah. So we have a Modal skill now that. Which is why we built this Modal Bench. It's to find things like that, so we can address them in our tool.Swyx [00:37:35]: Tune a skill. Yeah.Akshat [00:37:36]: Yeah.Swyx [00:37:36]: No. it's it's good. are you facing any shortages? like we talk a lot about GPU shortages, but also CPU, also memory.Swyx [00:37:44]: Yeah.Akshat [00:37:45]: We have had a lot of growth, which means that, there's - we've had to be much better aboutSwyx [00:37:53]: PlanningAkshat [00:37:54]: Proactive capacity planning.Swyx [00:37:55]: Yeah.Akshat [00:37:55]: So we have,Swyx [00:37:57]: Which by the way, like it's like a MBA's like dreamAkshat [00:38:00]: YesSwyx [00:38:00]: Is like just planning this stuff. I think last time you and I talked about something maybe about this.Akshat [00:38:03]: Yeah. we have a really competent team of people that we call, The role is called compute strategy. so yeah, if anyone listening here or wants to work on thatSwyx [00:38:13]: Compute strategy?Akshat [00:38:13]: Yeah.Swyx [00:38:14]: I think,Akshat [00:38:14]: I feel like,Swyx [00:38:15]: I think the normies call it FP&A or something.Akshat [00:38:18]: Well, it's more It's it's not FP&A. It's it's There's a lot of interesting financial questions of like what is the blend between one year and three-year reservations? how do we forecast our own capacity? how do we. especially since our capacity is very fungible across different GPU types and different regions, like you have to model a lot of it. and you also have to have an opinion on how the supply chain is gonna evolve, and then you have to like, take bets,Swyx [00:38:49]: YeahAkshat [00:38:49]: Based on that.Swyx [00:38:50]: Tokenomics.Akshat [00:38:50]: Yeah.Swyx [00:38:51]: This is like probably a not a real point, but, I was trying to think about like what other industries. I was trying to think about like, we cannot be first to like these kinds of problems.Akshat [00:38:59]: Yeah.Swyx [00:39:00]: And what other industries have had this? And I was like, airlines with fuel and like they have to hedge their fuel and like, I think for a long time Southwest because they made like a hero fuel bet, they like were like super low cost becauseAkshat [00:39:12]: OhSwyx [00:39:12]: Compared to everyone else.Akshat [00:39:14]: Yeah. I hadn't thought about that.Vibhu [00:39:16]: We're at a fun time too?Akshat [00:39:18]: Yeah. It's. A lot of the compute business in general, for us is also about being very good about capacity management. That is how you have great unit, economics. but also over time it's how you can unlock more value for customers. Like, one of the things we're building now is like a way for customers to get, If they don't care about latency, like get much cheaper pricing and they'll get results back in like next 24 hours or something, like a batch tier essentially.Batch Tiers and Latency-Insensitive WorkloadsSwyx [00:39:47]: Yeah.Akshat [00:39:47]: And those are levers we have because we control the whole stack and scheduling and whatnot to give people a sufficientSwyx [00:39:53]: Yeah. I feel like they're not as popular. Like those, like the Frontier Labs have all those APIs. They're not as popular as they should be.Akshat [00:40:00]: The demand that we see for something like that is not for LLMs. although sometimes people wanna run evals andSwyx [00:40:08]: OkayAkshat [00:40:08]: Synthetic data prep and there it makes sense.Swyx [00:40:10]: Okay.Akshat [00:40:11]: But it's from a lot of LLM companies, like people who are doing computational bio, like they have to run really big batch jobs and they don't care about when they get it back.Swyx [00:40:22]: Yeah. And like they have a reasonable. It's it's also like a cousin to the stopping problem of like, will this finish in time?Akshat [00:40:30]: Yeah. You can bound it.Swyx [00:40:33]: Yeah.Akshat [00:40:33]: Like you can give peopleSwyx [00:40:34]: YeahAkshat [00:40:34]: SLAs on it.Swyx [00:40:35]: Yeah. I think what's, what's interesting is like the next phase of model.Swyx [00:40:38]: Like what, do people expect from you, now that you're established and you're like well-known compute player among all these leading companies. You had an inference launch week, and we talked a little bit about the launches. like what else? Like what else should people know?What Modal Builds NextAkshat [00:40:55]: We are building primitives that make our users' lives much easier. So, I think for example, with LLM inference, thousands more companies are gonna post-train their own models and, deploy open source models for inference. so we're thinking a lot about what is the best product shape for that. And, that involves everything from our training gym to, then, endpoints that get frontier-level performance. again, but I haven't talked to anyone. It looks somewhat different on other verticals. Like, we're also seeing a lot of real-time, audio-video stuff in there, which is why like, we're working on things like regional routing, with fallbacks. So you can get GPUs that are as close to users as possible. so you get like low latency for video streaming and whatnot. And then on the agent side, it's,Akshat [00:41:52]: We're still working very closely with our customers because stuff is changing so fast in terms of what they need. And, I think beyond sandboxes and persistent file systems, there's a lot of other things people will need from this agent stack as they build production agents. So yeah, we're thinking about those other things that fit in there.Swyx [00:42:13]: I want to ask what the other things are.Akshat [00:42:15]: Yeah. I probably should share right now.Swyx [00:42:17]: I think-- I think, okay, so, I do think a lot about the principal components of cloud, and you do talk about compute storage networking.Akshat [00:42:25]: Yeah.Swyx [00:42:25]: Because so far for me, it's fine. so far for the. the first couple generations of cloud, it's fine. What's different, qualitatively different about agents that you need some new permission level? Like a lot of people, okay, and I'll just kinda spew tokens at you until it like hopefully sparks something.Akshat [00:42:43]: Yeah.Swyx [00:42:44]: Like the new level now is whatever Claude Code does, which is dangerously scope permissions or like allow list by command or like whatever, right? And sometimes they're like, “Well, okay, we have like this adaptive thinking mode where like, just trust me, bro. I will make the calls for you.” Is that it? like mediated permissions.Hard Guardrails vs. LLM-Mediated PermissionsVibhu [00:43:03]: Now you're looping it with a goal and letting it roll.Akshat [00:43:06]: Yeah, I'm, I'm skeptical of LLM media permission for stuff that is at the sandbox level because you do want hard boundaries.Swyx [00:43:16]: Yeah.Akshat [00:43:16]: Otherwise, someone can exfiltrate stuff.Swyx [00:43:20]: But likeAkshat [00:43:20]: YeahSwyx [00:43:20]: Maybe that's old school thinking. Maybe we're the dinosaurs.Swyx [00:43:23]: Maybe the AI OS or the LLM OS is really the kernel is a goddamn LLM.Swyx [00:43:30]: Like it makes you feel uncomfortable.Akshat [00:43:31]: Yeah, I'm, I'm toldSwyx [00:43:32]: But that's what trusting the LLM is. Like imagine a spherical cow perfect LLM.Akshat [00:43:36]: Right.Swyx [00:43:37]: That it.Akshat [00:43:39]: Maybe.Swyx [00:43:41]: I wanna test the boundaries, right?Akshat [00:43:42]: Yeah.Swyx [00:43:42]: Like, and I don't believe that, but I wanna see where I'm wrong ‘cause that's, that's the consensus.Akshat [00:43:49]: Yeah. I think you always need hard guardrails when you want, And you can pair those with softer guardrails, right? And that's gonna be a lot of mediated.Managed Agents and Specialized SandboxesSwyx [00:44:00]: There. I'll also get you a end with a couple of your commentary on like the ecosystem outside of Modal. Manage agents. Everyone has one. Gemini, OpenAI, Claude, very useful for you, but also like it is their way of starting to edge into your space.Akshat [00:44:17]: Yeah.Swyx [00:44:17]: What's going on?Akshat [00:44:19]: Yeah, we're, very excited to partner with Anthropic and some of the other foundation labs, will not name who we're also working with. the way we see it is the manage agent thing is a great place to start if you're starting out building an agent and, But then when you get to, building something more production grade, like you're a company that's like Ramp that's building their own, Ramp also runs their accounting agent on us, so their external-facing agent. You need a lot more control over, your compute primitive on things like, what sort - how do you persist different files that the agent has access to, and how do you snapshot and restore? How do you control the networking? maybe you want GPUs. When you get to that point, you kinda want, a specialized sandbox provider, that gives you those things, and that's the role that we are trying to play.Swyx [00:45:15]: YeahAkshat [00:45:16]: We don't really have an opinion on the harness, whether it runs - it's a cloud-managed agent, and you hook it up to Model Sandbox, or you run the harness in Model Sandbox. We'll see where people converge with that.Swyx [00:45:26]: Yeah. Do you any opinions on like the meta harnesses, or just another layer on top of these things?Akshat [00:45:31]: You mean like the OpenPipeSwyx [00:45:33]: OpenPipe is one. I think Vercel had one, which I can't remember the name of right now. Fredshot had one. and then, to me, most recently was Data Databricks that had Omnigen. All these are meta harness. Like it's kinda pseudo agent cloud type things.Akshat [00:45:50]: I personally have not played around with them.Swyx [00:45:53]: Yeah.Akshat [00:45:53]: Build agents with them.Swyx [00:45:54]: Everything's bullish Modal, as long as it consumes more infra.Akshat [00:45:57]: That's why we're focusing on the infra layer. It's somewhere where our, relative competence is and, also it's a hard problem to solve.Swyx [00:46:06]: Yeah. I will say like just generally reflecting on that, I don't know if - if there's other topics on Modal, but like just generally reflecting as an infra person, not as intense as you, but in that field, this has like been the most exciting time in infra. Like it was boring for a while, and you couldn't really get people excited about data infrastructure. Like Eric would get on Data Console, everyone just watched the video and like say, “Look at how many sandboxes I can spin up,” and no one gave a crap.Why Infrastructure Became Exciting AgainAkshat [00:46:39]: Yeah.Swyx [00:46:40]: And like now everyone gives a crap.Akshat [00:46:42]: That's true. It is a very exciting time, and I think a lot of that's driven by just the amount of scale all of this stuff needs.Swyx [00:46:50]: I think the, like a lot of your initiatives or a lot of your like product directions make sense in retrospect, which is like the best kind, but I wouldn't necessarily have thought about it myself, which.Akshat [00:47:00]: We need the predictions.Swyx [00:47:02]: I think there's a lot that you just don't even see, right? Like you have the batch, you have the voice, you have the multimodal, but what else?Akshat [00:47:10]: What else is coming up for usSwyx [00:47:11]: Yeah. Where do you see things going?Akshat [00:47:13]: Yeah. I, in generalBiotech, Robotics, and Non-LLM AI WorkloadsAkshat [00:47:15]: It's it's clear that there's there's a huge shift happening. I think one thing that's not as obvious to people because LLM inference gets talked about so much and is also we work a lot of companies that are, doing things like drug discovery and computational bio, like the Chai Discoveries of the world. Big things are probably gonna happen there. we work a lot of robotics companies that are putting robots in like active deployments and getting good results out of them.Swyx [00:47:45]: Is there Air Gap Modal? Is there a version that is like prem air gapped whatever?Akshat [00:47:50]: No. We,Swyx [00:47:51]: You should cloud only.Akshat [00:47:51]: Yeah.Swyx [00:47:52]: Yeah. Okay. But yeah, so what you're saying is like because you're focused on primitives and they're good primitives, you find use cases in all these kinds of things.Akshat [00:48:01]: Yeah.Swyx [00:48:01]: Probably diversifies you a little bit away from LMS all the time.Akshat [00:48:05]: Yeah, absolutely. We're, we'- our goal isn't to only serve the LLM inference market.Swyx [00:48:10]: There are a lot just on the website, the audio,Akshat [00:48:12]: Yeah. We said both onSwyx [00:48:14]: Computational bio images. Yeah, there's a lot here. There's QTA TTS, customizing. Oh, Chatterbox. there was customizing Whisper.Akshat [00:48:24]: Okay. Yeah.Swyx [00:48:25]: This screen reminds me of a fallen competitor, which Replicate.Model APIs vs. Differentiated AI ProductsSwyx [00:48:31]: What's your postmortem on what happened?Akshat [00:48:34]: This is one thing we've stayed away from is providing an API for models because I think providing model APIs is some of it ends up serving like a really hobbyist market, which is much less sticky.Swyx [00:48:50]: Yeah.Akshat [00:48:50]: And we've always wanted to build for companies that are building products and need more flexibility that's not just an API.Swyx [00:48:57]: Which you can build an API for a model and this is clearly what it is. But you - but what you're saying, you can wrap it into a more fully functioning back end that you run.Akshat [00:49:06]: Yeah. So all of our examples, it's not that spin up this model, here's an API token, use it. They're all code.Swyx [00:49:13]: Okay.Akshat [00:49:13]: And so the point is that this is just an example.Swyx [00:49:16]: Starter code.Akshat [00:49:17]: Yeah. But you can tweak it however you want.Swyx [00:49:20]: Yeah.Akshat [00:49:21]: And if you're like a company building a product, like, computational bio whatnot, yeah.Swyx [00:49:26]: I guess I'm trying to tease out for listenersAkshat [00:49:28]: YeahSwyx [00:49:28]: When does it stop becoming, oh, you're just an API call and you're just a wrapper on API to becoming what you call a product, right?Swyx [00:49:36]: Like, what is that layer? Like what-- Like, more lines of code, but like beyond that, what is the substance that people add that qualifies it to be something more?Akshat [00:49:46]: I think there's a little bit of like a selection effect of like a lot of the companies who do wanna get deeper into that level are probably building something that's more differentiated. And, I think, an example is like - with LLM inference, originally we, worked with companies that were building their own post-training frameworks or they were, - Ramp early in the day was training their own tokenizer and like swapping out the tokenizer in Llama and whatnot. I'm not saying that's, that successful, in that case. But a better example is like, let's say Suno. because Suno, does not use Modal for training.Swyx [00:50:26]: Mikey on the pod. Yeah.Akshat [00:50:27]: But they use Modal for all their inference and that's because they have like a custom-- They have completely custom model architecture and that means that they have to be at the code level and tweak things that are not, just an API.Swyx [00:50:41]: It's interesting as well, like we had, Ethan, most recently on the xAI Groq team make a prediction that like the next tier in video gen is not a better video model, it's a better model or agent that orchestrates video models.Video Agents and Production WorkflowsAkshat [00:50:56]: Oh, interesting.Vibhu [00:50:56]: Language model backbone that can use toolsAkshat [00:50:58]: RightVibhu [00:50:59]: And write code.Akshat [00:51:00]: Like, yes, I can make my second video or my second video from Groq, but I want my minute video.Akshat [00:51:06]: And I'm not going there through normal video gen.Swyx [00:51:10]: Yeah, that's interesting. I - So we have GPU sandboxes and recently have seen a few companies doing agents that do video manipulation or,Akshat [00:51:22]: Yeah. Give it FFmpeg and just do it.Swyx [00:51:23]: Run FFmpeg. But likeAkshat [00:51:25]: That's not enough.Swyx [00:51:25]: Yeah.Akshat [00:51:26]: You need to give it Adobe.Swyx [00:51:27]: Yeah, I hadn't put it together with like it would be a video production thing. in my mind these things were going more towards editingAkshat [00:51:36]: Yeah.Vibhu [00:51:36]: Well, shout out Mantis.Akshat [00:51:37]: I think about this a lot.Swyx [00:51:38]: .Akshat [00:51:41]: Yeah. Sorry.Vibhu [00:51:41]: Luma. Luma Agent is a version of this for video production, but it's a off.Swyx [00:51:46]: I was gonna get your quick takes, on some other stuff that happensGitpod/Ona, CI, and Runtime SandboxesSwyx [00:51:50]: In recent news and just-just see if you have anything interesting. Gitpod, very li

KI-Souveränität: Wenn dein Spitzenmodell über Nacht abgeschaltet wird   Ein Regierungs-Kill-Switch für dein wichtigstes KI-Modell ist keine Zukunftsmusik mehr, sondern gerade Realität geworden. Fable 5 stand nur 48 Stunden uneingeschränkt zur Verfügung, bevor es außerhalb der USA über Nacht wieder abgeschaltet wurde.   Torsten Koerting auf LinkedIn: LinkedIn - https://www.linkedin.com/in/torstenkoerting/   Warum das ein Souveränitätsproblem ist Wenn eine Regierungsentscheidung ausreicht, um ein Modell über Nacht vom Markt zu nehmen, hast du als Unternehmer die Kontrolle über deine eigenen Prozesse verloren. Torsten Koerting nennt das offen ein Erpressungsrisiko für Europa und für jedes Unternehmen, das kritische Abläufe auf ein solches Modell aufbaut. Wer produktive Geschäftsprozesse an eine Technologie bindet, die per Kill Switch abgestellt werden kann, betreibt damit im Kern ein Glücksspiel. Erst wenn eine Technologie wirklich demokratisiert und uneingeschränkt zugänglich ist, lässt sie sich verantwortungsvoll produktiv einsetzen. Testen ja, produktiv einsetzen nein Solange die Rahmenbedingungen für stabilen Zugriff fehlen, gehören Spitzenmodelle wie Fable 5 in Testumgebungen, Playgrounds und Sandboxes, nicht in kritische Geschäftsprozesse. Für das Tagesgeschäft reicht in den meisten Fällen ohnehin ein niedrigschwelligeres Modell, etwa für Blogbeiträge, News Reports oder Chat-Anwendungen. Erst wenn du KI nutzt, um wirklich etwas völlig Neues entstehen zu lassen, wird das teuerste und leistungsstärkste Modell überhaupt relevant. Diese Unterscheidung entscheidet darüber, wo Abhängigkeit von einem einzelnen Anbieter riskant wird und wo nicht. Die vier Formen der KI im Unternehmen Aus dieser Erfahrung ist bei Koerting Institute ein neues Einordnungsmodell entstanden: integrierte, additive, Unternehmens- und disruptive KI, verortet zwischen den Achsen generisch bis individuell und evolutionär bis disruptiv. Es hilft dir einzuordnen, ob du gerade mit der KI arbeitest, sie also wie ein Werkzeug im Tagesgeschäft nutzt, oder durch die KI etwas völlig Neues schaffst. Diese Einordnung entstand erst in den letzten Wochen aus einem Impuls in einem Corporate Workshop und wird laufend weiterentwickelt. Fazit Die Fable-5-Episode zeigt sehr konkret, dass technologische Abhängigkeit von einem einzelnen, politisch beeinflussbaren Modell ein echtes unternehmerisches Risiko ist. Prüfe deshalb, welche deiner Prozesse wirklich auf ein Spitzenmodell angewiesen sind und welche mit einem stabileren, austauschbaren Modell genauso gut laufen.     Noch mehr von den Koertings ...  Das KI-Café ... jede Woche Mittwoch (>350 Teilnehmer) von 08:30 bis 10:00 Uhr ... online via Zoom .. kostenlos und nicht umsonstJede Woche Mittwoch um 08:30 Uhr öffnet das KI-Café seine Online-Pforten ... wir lösen KI-Anwendungsfälle live auf der Bühne ... moderieren Expertenpanel zu speziellen Themen (bspw. KI im Recruiting ... KI in der Qualitätssicherung ... KI im Projektmanagement ... und vieles mehr) ... ordnen die neuen Entwicklungen in der KI-Welt ein und geben einen Ausblick ... und laden Experten ein für spezielle Themen ... und gehen auch mal in die Tiefe und durchdringen bestimmte Bereiche ganz konkret ... alles für dein Weiterkommen. Melde dich kostenfrei an ... www.koerting-institute.com/ki-cafe/   Mit jedem Prompt ein WOW! ... für Selbstständige und Unternehmer Ein klarer Leitfaden für Unternehmer, Selbstständige und Entscheider, die Künstliche Intelligenz nicht nur verstehen, sondern wirksam einsetzen wollen. Dieses Buch zeigt dir, wie du relevante KI-Anwendungsfälle erkennst und die KI als echten Sparringspartner nutzt, um diese Realität werden zu lassen. Praxisnah, mit echten Beispielen und vollständig umsetzungsorientiert. Das Buch ist ein Geschenk, nur Versandkosten von 9,95 € fallen an. Perfekt für Anfänger und Fortgeschrittene, die mit KI ihr Potenzial ausschöpfen möchten. Das Buch in deinen Briefkasten ... https://koerting-institute.com/shop/buch-mit-jedem-prompt-ein-wow/   Die KI-Lounge ... unsere Community für den Einstieg in die KI (>2800 Mitglieder) Die KI-Lounge ist eine Community für alle, die mehr über generative KI erfahren und anwenden möchten. Mitglieder erhalten exklusive monatliche KI-Updates, Experten-Interviews, Vorträge des KI-Speaker-Slams, KI-Café-Aufzeichnungen und einen 3-stündigen ChatGPT-Kurs. Tausche dich mit über 2800 KI-Enthusiasten aus, stelle Fragen und starte durch. Initiiert von Torsten & Birgit Koerting, bietet die KI-Lounge Orientierung und Inspiration für den Einstieg in die KI-Revolution. Hier findet der Austausch statt ... www.koerting-institute.com/ki-lounge/   Starte mit uns in die 1:1 Zusammenarbeit Wenn du direkt mit uns arbeiten und KI in deinem Business integrieren möchtest, buche dir einen Termin für ein persönliches Gespräch. Gemeinsam finden wir Antworten auf deine Fragen und finden heraus, wie wir dich unterstützen können. Klicke hier, um einen Termin zu buchen und deine Fragen zu klären. Buche dir jetzt deinen Termin mit uns ... www.koerting-institute.com/termin/   Weitere Impulse im Netflix Stil ... Wenn du auf der Suche nach weiteren spannenden Impulsen für deine Selbstständigkeit bist, dann gehe jetzt auf unsere Impulseseite und lass die zahlreichen spannenden Impulse auf dich wirken. Inspiration pur ... www.koerting-institute.com/impulse/   Die Koertings auf die Ohren ... Wenn dir diese Podcastfolge gefallen hat, dann höre dir jetzt noch weitere informative und spannende Folgen an ... über 500 Folgen findest du hier ... www.koerting-institute.com/podcast/   Wir freuen uns darauf, dich auf deinem Weg zu begleiten!

Have you checked out Azure Container Apps lately? Carl and Richard talk to Paul Yuknewicz and Jan Kalis about how they have sped up startup time with Container Apps using a feature called Express, and then on to the big news - Sandboxes. Sandboxes let you experiment with Azure Container Apps with managed identity to access Azure Connectors to all sorts of resources, both within Azure and outside it - Salesforce, SAP, etc. It's an effective and secure strategy for building applications, but also for experimenting with agents - at no point does the agent actually have control of any credentials. If you haven't looked at ACA lately, you need to check out these latest features!

Have you checked out Azure Container Apps lately? Carl and Richard talk to Paul Yuknewicz and Jan Kalis about how they have sped up startup time with Container Apps using a feature called Express, and then on to the big news - Sandboxes. Sandboxes let you experiment with Azure Container Apps with managed identity to access Azure Connectors to all sorts of resources, both within Azure and outside it - Salesforce, SAP, etc. It's an effective and secure strategy for building applications, but also for experimenting with agents - at no point does the agent actually have control of any credentials. If you haven't looked at ACA lately, you need to check out these latest features!

Saket Narayan of AWS joins Jo Ann to explore how regulators can move beyond pilots and sandboxes, using cloud, data, AI, and secure experimentation to build faster, smarter, more adaptive supervision. Episode show notes, transcript and related links available online at: https://regulationinnovation.org/podcast/beyond-pilots-sandboxes/.

We're excited to have Databricks join us at AIEWF, among hundreds of the top companies in the AI Engineer ecosystem. LS subscribers can use their discount to get past the late bird pricing and access over $50k in sponsor offers! Everyone is still talking about Satya's Frontier Ecosystems post, but few have actually built a (now $175 billion) frontier ecosystem and cloud like our guests today.From open-sourcing the layer above coding agents to rethinking databases for the agent era, Databricks cofounders Matei Zaharia and Reynold Xin are pushing the company beyond the lakehouse into a full data-and-AI operating system. In this episode, Matei and Reynold join swyx at the 2026 Data + AI Summit to unpack Omnigent, LTAP, Lakebase, agent security, open formats, Mosaic, and why databases may matter more than ever once AI agents start doing real work.We go deep on Omnigent: Databricks' open-source meta-harness for combining, controlling, and sharing agents across Claude Code, Codex, Cursor, Pi, custom agents, and internal tools. Matei explains why coding agents and enterprise agents run into the same problems: portability, collaboration, session history, security, spend controls, and the need for a common API above every harness.Then Reynold walks through Databricks' database dream: why CDC is brittle enough to joke that it means “continuous data corruption,” why HTAP has been the holy grail of database engineering, and why Databricks thinks LTAP gets most of the benefits by unifying the storage layer instead of collapsing every query engine. We also cover Databricks' infrastructure scale, the culture behind rapid prototyping, the difference between tech and enterprise customers, Databricks vs Snowflake, whether vector databases should have ever existed, the Mosaic model strategy, Genie, AI Runtime, RL fine-tuning, and the thesis that traditional software gets rewritten once the data is in the right place and agents sit on top.Databricks began as a company for the big data era. The origination of Spark from the Berkeley AMPLab which eventually turned into the product Lakehouse convinced enterprises that they didn't need a separate data lake, warehouse, ML platform, and governance layer. They just needed one open foundation where all of their data could live and be reasoned over.Since then a lot has changed, but data has only become more important. Data is no longer something you keep track of and analyze ad hoc, it's the necessary context agents need in order to act. So the framing has shifted from “where do we put all of our data?” to “how do we expose the right slice of state, history, permissions, and business logic to an AI system at the exact moment it's doing work?”If frontier model performance becomes commoditized, the durable advantage then becomes the company-specific context around them: proprietary data, governed access, operational state, transaction logs, workflows, and feedback loops. Which makes Databricks positioned perfectly.Now coming fresh off the Data + AI Summit 2026, the company is moving just as fast to keep up, announcing Genie One, Omnigent, LTAP, and many more, indicating a central mission in its newer work: Databricks is trying to become the operating system for enterprise agents.Models are getting good enough, but agents are only useful if they have the right context, permissions, memory, state, cost controls, and access to live business data. Fundamentally it appears that significantly better model performance in production is a systems problem, one that data guys like us are remarkably well prepared to solve!We discuss:* Why Databricks built Omnigent as a meta-harness above existing AI agents* Why coding agents and custom enterprise agents need the same infrastructure* The common API for agent sessions, files, streams, tool calls, and cancellation* Why persistent sessions, cloud sandboxes, sharing, search, and collaboration matter* Why Databricks open-sourced Omnigent instead of keeping it proprietary* Databricks' internal agent usage, cloud sandboxes, and coding workflows* The scale of Databricks: 50–60 million virtual machines a day and exabytes before breakfast* Why agent security needs contextual and stateful policies* How an agent could read confidential docs, install a compromised npm package, and leak data* Why spend control matters when an agent can burn $500 reading logs* Startup opportunities around coding-agent analytics, quality, skills, and spend* LTAP, Lakebase, and why Databricks wants to rethink the database stack* OLTP vs OLAP, CDC, and why data pipelines break at 3 a.m.* Why HTAP has historically been the holy grail of database engineering* Why Databricks thinks LTAP is “HTAP done right”* How writing transactional data into column-oriented formats changes analytics* Why agents need live operational context from databases, not just telemetry* How Databricks prototypes strategic systems without endless process* Enterprise vs tech customers, governance, procurement, and DIY culture* The “second system syndrome” risk of rewriting a database engine* Building a database engine from a decade of traces and quadrillions of data points* Why vector databases should never have been a separate category* Why open formats and AI changed the race with Snowflake* The Mosaic story, DBRX, Genie, document parsing models, and specialized model training* Why model customization and RL fine-tuning may become mainstream* Why “get the data there, slap some agent on top” may rewrite traditional softwareMatei Zaharia* LinkedIn: https://www.linkedin.com/in/mateizaharia* X: https://x.com/matei_zahariaReynold Xin* LinkedIn: https://www.linkedin.com/in/rxin* X: https://x.com/rxinDatabricks* Website: https://www.databricks.com* X: https://x.com/databricksTimestamps00:00:00 Introduction00:02:22 Omnigent and the Agent Infrastructure Layer00:08:39 Agent Clouds, Common APIs, and Open Source00:16:52 Databricks Scale and Internal AI Workflows00:18:03 Agent Security, Governance, and Spend Controls00:27:34 LTAP and the Database Dream00:30:30 CDC, HTAP, and Why Data Pipelines Break00:34:05 Lakebase, Parquet, and Live Data for Agents00:36:47 Databricks' Culture of Fast Prototyping00:43:40 The Dream Engine and Rewriting the Database Stack00:51:02 Vector Databases, Query Engines, and LTAP00:52:36 Databricks vs Snowflake00:57:48 Mosaic, DBRX, Genie, and Specialized Models01:03:11 Context, AI Runtime, and RL Fine-Tuning01:06:15 Why Data + Agents May Rewrite Software01:07:09 Closing ThoughtsTranscriptIntroduction: Databricks, Data + AI Summit, and Founder DynamicsSwyx [00:00:00]: Matei and Reynold from Databricks, welcome to Latent Space.Reynold Xin [00:00:06]: Hey, thanks for having us.Swyx [00:00:07]: Yeah.Matei Zaharia [00:00:08]: Yeah, thanks so much.Swyx [00:00:09]: thanks for taking time out. You have your Databricks, Data AI Summit going on. You were just telling me how the first summit that you guys ran was just 50 peopleReynold Xin [00:00:17]: Yeah, it wasSwyx [00:00:17]: in BerkeleyReynold Xin [00:00:18]: little meetup at Berkeley, I thinkMatei Zaharia [00:00:19]: YeahReynold Xin [00:00:19]: put togetherMatei Zaharia [00:00:20]: We were doing these tutorials and, yeah, just teach people Spark.Swyx [00:00:23]: Yeah. obviously now it's like, I think like the headline number's like 100,000 people around the world, 30,000 in person.Swyx [00:00:30]: it's a crazyMatei Zaharia [00:00:31]: AmazingSwyx [00:00:31]: community. Well, I just saw the keynote.Swyx [00:00:35]: Ali's just. Did was it obvious or that back when that Ali would be, like, such a great, like, CEO? LikeReynold Xin [00:00:42]: OhSwyx [00:00:42]: such a great presenter?Reynold Xin [00:00:43]: What do you think?Matei Zaharia [00:00:44]: I think among our group of founders it was clear that, I think he'd be the best at this.Swyx [00:00:50]: Yeah.Matei Zaharia [00:00:50]: And yeah, it turned out great. And he's, he's ramped up on so many topics growing a company. He would just go in and, like, study it and, be talk to all the experts. Like, even if he can't hire the person, learn enough about, like, finance and sales and whatever it was, and, and go from there. Yeah.Swyx [00:01:09]: Yeah.Reynold Xin [00:01:10]: he's obviously very high IQ and a very high EQ, but it wasn't. Like, Ali today is quite different from Ali from, like 10 years ago. I think there's a lot of work that he put in to, get to this point.Swyx [00:01:20]: Yeah. no, to me the most appealing thing about him is that he's funny. And like, it, it's, it'Matei Zaharia [00:01:26]: It's true, yeahSwyx [00:01:26]: it's hard to make jokes about, data warehousesReynold Xin [00:01:30]: About serious topicsSwyx [00:01:31]: securityMatei Zaharia [00:01:32]: YeahSwyx [00:01:32]: what have you.Matei Zaharia [00:01:33]: Oh, yeah. That's for sure.Swyx [00:01:34]: Yeah. So you guys launched a whole bunch of things. I'll, I'll just name check briefly, the stuff because we're not gonna cover everything. Omnigentt, your baby. LTAP, your baby, your dream engine.Swyx [00:01:47]: we're also gonna cover Genie, cover CustomerLake, you acquired PantherMatei Zaharia [00:01:52]: YeahSwyx [00:01:52]: Open Sharing, and there's Unity AI Gateway. A lot of these, I think, like, are things that you would expect a Databricks to do. It's, it's like part of the roadmap. Everyone in your category has similar things. But I think, probably the two of you are leading the two most unique and differentiated initiativesOmnigent and the Agent Infrastructure LayerSwyx [00:02:09]: on, in the landscape. Maybe we'll start with, Omnigentt we'll, we'll, we'll, we'll go into it. I do think that a lot of people are exploring this meta harness concept.Matei Zaharia [00:02:21]: Yeah, totally.Swyx [00:02:21]: What led you to it?Matei Zaharia [00:02:22]: Yeah. There were a couple of, like, converging lines, which I think is a good sign that you need something new. So on the one hand, there's all the coding agent info internally. We have really great, dev infra team. they built something called Isaac, that's like a wrapper on Claude Code and Codex, and, lets you use them either on the web in, like, sandboxes or, just on your dev machine or on your laptop or whatever. And then, they were adding all kinds of stuff there. And we saw all the more advanced engineers like, were building their own workflows with tons of agents, and they were building their own UIs and stuff on top or even on top of that. And then the other one was, like, us building agents. We ship this, like, data science agent called Genie on the research team, which I lead. We also build a lot of internal ones for various things, and then we have all the customer ones. And all of them running into this thing of like, “Oh, I need to switch model and harness and so on,” every few months. Plus the agent is, like, completely useless if you can't share sessions with someone and have history and have search and all this, like, layer on top of it for collaboration. I thought a bit about it from both contexts and, at first people thought it was weird. They're like, “Why are you doing coding agents and custom agents in the same thing?” But I said it's, it's the same problems and, you just wanna build the stuff that lets you deliver the agent, maybe control it if you care about security, and, make it portable across things. And then we prototyped some things as experiments. We saw, yeah, we can make it work, and then we built that for real.Swyx [00:04:06]: I'm wondering if this let's call it architectureMatei Zaharia [00:04:11]: YeahSwyx [00:04:11]: maps to anything in your careers in the past. like I always think about how a lot of things just tie back to operating systems.Swyx [00:04:18]: A lot of operatingMatei Zaharia [00:04:19]: YeahSwyx [00:04:20]: systems tie back to databases,Matei Zaharia [00:04:21]: SoSwyx [00:04:21]: or the other way aroundMatei Zaharia [00:04:22]: so the thing, I do think it ties a lot to, like, network protocols, internet protocol. we alsoSwyx [00:04:29]: Communication between entities.Matei Zaharia [00:04:30]: Yeah. We did stuff with, like, data sharing also, which is probably, most viewers probably won't know unless they'Swyx [00:04:36]: Yeah, open protocol is the term.Matei Zaharia [00:04:37]: Yeah.Swyx [00:04:38]: Open sharing. Open sharing.Matei Zaharia [00:04:38]: Open sharing.Swyx [00:04:39]: Yes.Matei Zaharia [00:04:39]: Yeah. So it's like you have a company, you maintain some table, like let's say like a Walmart or something. They have like the, inventory and what's been sold in each store. And then you also have suppliers, and they would love to produce more things and ship them, like, exactly the moment you need them. So they would love, like, real-time access to your table. So instead of like sending emails around or Excel sheets or phone calls, why can't you share like a view of that table in real time with them? Then they query, they, join it with their data, and they decide what to send. So it's one of these things where you, like you might ask like today since we can vibe code anything so fast, why do we even need to design like protocols or APIs or software? Why can't you just vibe code things on demand? But for this type of interoperability where multiple parties that are moving at different speeds are building stuff and you still want some layer on top to coordinate, you do wanna design it and build it. So it reminds me of that, like agents talking to each other and, users talking to agents and tools.Agent Clouds, Cloud Sandboxes, and Keeping Sessions AliveSwyx [00:05:42]: Reynold, any other comments alternative viewpoints?Reynold Xin [00:05:46]: I think, by the way, we had a debate on exactly which set of benefits would, matter a lot, and I think around the time we decided to do this thing I was telling Matei, “Hey,” it just happened to be there's a particular week that I was coding nonstopSwyx [00:06:00]: from the moment I woke up to, like, the moment I went to bed, I was, like, looking at my Claude sessions, my Codex sessions. And one of the things that was particularly annoying was having to keep my laptop open.Swyx [00:06:12]: I was driving to a doctor's appointment, and I remember because I wanted to make sure the whole thing continues working.Matei Zaharia [00:06:18]: But by the way, it's so comforting to hear you say that because I'm like, “I don't know if I'm a clown and I'm doing this or like.”Swyx [00:06:25]: Yeah. Like honestly, I was driving and I was tethering my laptop to my phone.Matei Zaharia [00:06:29]: huh.Swyx [00:06:29]: Keeping it on the side. Whenever I hit a red light, I started looking at what's going on my laptop.Matei Zaharia [00:06:35]: Yeah.Swyx [00:06:35]: And I just felt that was ridiculous.Matei Zaharia [00:06:37]: Yeah.Swyx [00:06:37]: It felt like we went back to the dark agesMatei Zaharia [00:06:39]: YeahSwyx [00:06:40]: programming. the productivity you gain from all this coding age is amazing, but, yeah.Matei Zaharia [00:06:45]: Have you heard of cloud?Swyx [00:06:47]: Yeah.Swyx [00:06:48]: It was crazy to me.Matei Zaharia [00:06:49]: Oh, the thing you were working on was the sandboxes or was this before that?Swyx [00:06:52]: It was a sandbox.Matei Zaharia [00:06:53]: Okay.Swyx [00:06:54]: I was workMatei Zaharia [00:06:54]: So you were inSwyx [00:06:55]: So I was approaching from a very different angle. I wanted to, “Hey, we're gonna have cloud sandboxes that doesn't shut down. You can get one very quickly,” but not just for running agentic sessions.Matei Zaharia [00:07:06]: Yeah.Swyx [00:07:06]: It's also for running development. So I was personally building that week, and through building that, I ran into all these issues, and then I wroteMatei Zaharia [00:07:15]: YeahSwyx [00:07:15]: a document for Matei, it's like, “Here's my wish list of what the actual environment should do.” And I think he ended up almost implementingMatei Zaharia [00:07:22]: YeahSwyx [00:07:22]: every single one of them.Matei Zaharia [00:07:23]: Yeah, I remember Reynolds saying, ‘cause my first prototype of this had just chats with your agent and he said, “I have to be able to open a shell, like my own shell and like list files and like tail them and stuff.” SoSwyx [00:07:36]: So SSH into a mainframe.Matei Zaharia [00:07:37]: Yeah. it has that now.Swyx [00:07:39]: Tailing my log.Matei Zaharia [00:07:40]: Yeah.Matei Zaharia [00:07:41]: Yeah.Swyx [00:07:41]: And also another thing I think I asked was, I had. I still use cursor for the sole purpose of rendering markdown files.Matei Zaharia [00:07:48]: huh. Yes.Swyx [00:07:49]: So I said, “If you just give me a way to see my markdown files and renderMatei Zaharia [00:07:53]: YeahSwyx [00:07:53]: them properly, I don't need a separate tool anymore.”Matei Zaharia [00:07:55]: Yeah.Swyx [00:07:56]: And I think you also built that in.Matei Zaharia [00:07:57]: Yeah, we, yeah, we did that, yeah. Yeah, we had a lot of engineers building, their own vibe coding setup. But then the other thing they all said is like, “Hey, I built something that's amazing for me, but, like, no one else on the team can use it ‘cause I don't have a server to collaborate.” And this is why we tried to set up, Omnigent, so you can have a server and have the security, set up in there. So, like log in with Google or whatever and, like securely share stuff. which. And that's where we've seen a lot of other agents like hit things. Like people think they prototyped an awesome agent, but it's not allowed to connect to like some really important data or whatever because of the security team.Omnigent Architecture, Open Source, and Common APIsSwyx [00:08:38]: Yeah.Matei Zaharia [00:08:38]: So yeah.Swyx [00:08:39]: Yeah. At this point, so for those watching along on YouTube, we're gonna putting up a image of the structure here, and we can talk a little bit of the architecture. I think I just want to have people understand, ‘cause like when we're talking about software, it can be very abstract and like here is what we're talking about. You've worked out in open source this entire platform and there's a runner component and server component with a uniform API that you've, you've figured out. any other element and obviously you can plug in all this, persistence layers and compute layers. This is a whole cloud. It's an agent cloud.Matei Zaharia [00:09:12]: Yeah. It's, it's got these components to work with it. The, a lot of the action happens like on the machine where you deploy your agent too. So whatever you've got on there, you can run. But yeah, it's, I think it's the minimal thing you want to have hosted, like collaborative agents and to have that server. And one of the reasons we open sourced it is, anyone building agents, this gives them an app they can start with and customize, which we were seeing in Databricks too. Like someone would make a nice, agent app and then other teams would ask, “Oh, can I just use yours for my agent?”Swyx [00:09:45]: Yeah, I think we had like five or six different agentic frameworksMatei Zaharia [00:09:48]: YeahSwyx [00:09:48]: built by every different team. They do all do more or less the same thing. Yeah, you need to. people wanna take something that works in Forkit, and you might as well have something open source. Yeah, which also was another question, which is interesting for Databricks. Like what do you choose to open source? What do you choose to make it proprietary? It's in. this goes back to Spark, right?Matei Zaharia [00:10:05]: Yeah.Matei Zaharia [00:10:06]: One, so one of the reasons to open source something is if you think it's a layer that will there'll be some network effect, it'll benefit from many, people collaborating, on it. So, for example, with Spark, I don't know if when Spark came out, we also focused a lot on letting you have libraries on top. So like there used to be differentSwyx [00:10:28]: EcosystemMatei Zaharia [00:10:28]: distributed computing engines for like machine learning and graph computation. We said they should all be libraries that you can compose. And we made it super easy to add connectors to data sources too. And then we benefit because, we don't have the time to write like connectors to like, 1,000 like different databases and file formats, but we can just use the ones people make, and of course they benefit from joining, this thing. So that's like one of these as it. Another way to think about it is like imagine, we our thing wasn't open. We had some agent hosting thing, but it's not open and then there is an open one. if you're. Which one's gonna win in the long run? So like here, because there is this benefit from like people writing integrations, it'll be, it'll be that. And then there are other things that like you just can't, even deliver as open source that are things the company does. Like for example, how do you make sure you're like streaming, jobs or your Lakebase database doesn't like, lose all your data at night? Well, that requires an operational team that's gonna sit there. There's no way it has to be a service. So like we wanna make sure as a company we're really good at those infra services and then we're as open as we can in terms of like what you build on top.Swyx [00:11:42]: speaking from a benefits, I think we are already seeing pull requestsMatei Zaharia [00:11:45]: YeahSwyx [00:11:45]: of all kinds of ecosystem integration, even though it was only released on Saturday.Matei Zaharia [00:11:50]: Yeah, Saturday. Yeah. So someoneSwyx [00:11:51]: Let's see, let's see what's going on. Yeah, you can look at the merge ones. I asked Sam Nigon this morning aboutMatei Zaharia [00:11:59]: 400 merge already?Matei Zaharia [00:12:00]: Yeah. I think Recent quite, I would guess around half are not from our team. but for example, someone added support for running it on Kubernetesrnetes. people added, many cloud sandboxes, so this can launch a cloud sandbox and run your agent in there, which is great for sharing too, ‘cause it's not, like, on your laptop and someone's, like, running scary code on there. so yeah, many startups have put those in, and, we expect to see more of them. We also have more agent harnesses already. Cursor, CLI, and Antigravity also.The Modern Data Stack and the Emerging AI StackMatei Zaharia [00:12:34]: Yeah. That's all, beautiful. And I, I feel like the last time this happens, there was the rise of the modern data stack.Matei Zaharia [00:12:42]: I don't know if it's that useful. I'm, I'm curious in your postmortem.Matei Zaharia [00:12:46]: I think most peopleSwyx [00:12:47]: AgreeMatei Zaharia [00:12:47]: will agree that it is finally dead. but maybe this arises to a new modern AI stack that, like, does the same thing.Matei Zaharia [00:12:52]: I don't know.Reynold Xin [00:12:54]: I think the modern data stack was a pretty useful thing, probably even up until this day. I think what, maybe for the audience who don't understand the history, I think the modern data stack is effectively decomposed into you need a layer to ingest the data in, you need a layer to transform your data, and then all of this are run, and then you need a layer to maybe visualize your data. And all of this runs on some data warehouse, or later on, as we're doing data warehouse or lakehouse.Reynold Xin [00:13:21]: I think that concepts are all very powerful and very useful. They enable a lot of workloads. What people eventually run into is a question of unification and consolidation is, hey, do you really need to chop all this into different pieces and work with so many different vendors and platforms in order to get, like, a very simple visualization done, right? So I think, like, over time, everybody started realizing that customers are pushing us. We started, we can realize that, so we started building more and more capabilities and trying to consolidate. And at the end of the day now, customers don't have to worry about having me hook up five different systems in orderMatei Zaharia [00:13:55]: YeahReynold Xin [00:13:55]: produce a chart. But the. I think, honestly, something like this is probably happening, in how many different frameworks do you want to hook up together in order to produce, like do a very simple agent.Matei Zaharia [00:14:06]: Just to be clear, I would say the core of this is this common API on top of all the harnesses. So the API is like, you've got an agent session, and you can send in a message or, like, a file. That's what you can send in, and then you get out, these streams as it's streaming text or as it's doing tool calls. And, or the other thing you can send in is you can, like, tell it to cancel a turn. So that's the API. Now, the thing we did is we could get you that on top of, like, cloud code running in a terminal, Codex, Py, OpenAI SDK, all that stuff. We map them all to that same interface. So that is something that you'd have to maintain yourself if you built your own, like, agent orchestrator, and then whenever cloud changes its API, you gotta, tweak your thing or it's gonna lose some messages. So that's the thing that's valuable to maintain. Then on top of that, like, we built a few apps. I think we built a pretty cool UI and stuff, but that's, And we built a security and control piece, which I'm excited about. But it's that common interface, so we don't. We. That doesn't try to be a stack. And in fact, you could plug in your own UI on top of this, server. That, and that's one of the use cases we care a lot about, ‘cause we want to use this in our own products.Compute, Sandboxes, and Databricks ScaleSwyx [00:15:20]: Yeah. It should be everywhere.Matei Zaharia [00:15:22]: Yeah.Swyx [00:15:22]: I think one of those things that is really interesting to me is, like, well, first of all, I'll, I'll endeavor to do everything and not call it the modern AI stack because like it needs a different name.Matei Zaharia [00:15:32]: Yeah.Swyx [00:15:32]: But like, yes, like, so one of the first people that told me about compute, sandboxing was Nikita from Neon.Swyx [00:15:39]: Because a lot of people think about Neon as like, well, it's serverless Postgres with, like, the separation of compute and storage and, instant branching and all those things. But every database company is also a compute company.Matei Zaharia [00:15:51]: Yeah. Yeah.Swyx [00:15:52]: And so he was showing to me his whole, his sandboxing solution. I don't think he have ever launched it.Matei Zaharia [00:15:57]: So our sandbox solution, the reason we could build it so quickly was because we realized if you just take the actual Lakebase architectureSwyx [00:16:05]: YeahMatei Zaharia [00:16:05]: and remove the database from it, by the coming from NeonSwyx [00:16:08]: Exactly, rightMatei Zaharia [00:16:09]: you have this sandboxSwyx [00:16:09]: Every database company has it already, yeah.Matei Zaharia [00:16:11]: Now, there are some differences. For example, in the one to support this particular workflow, it's important to have local persistence,Swyx [00:16:19]: YeahMatei Zaharia [00:16:19]: because you want your state to persist. Your libraries, you don't have to install your library every time, right?Matei Zaharia [00:16:24]: whereas the Neon architecture, because of the separation of storage from compute, you don't need persistent local disk.Swyx [00:16:30]: Yeah.Matei Zaharia [00:16:30]: So there's some differences.Swyx [00:16:32]: Yeah.Matei Zaharia [00:16:32]: But the, at the end of the day, yeah, it's, Yeah, so this is when you run, like, a coding sandbox. Like, if I use it, yeah, we have the dev env internally at Databricks. There's, like, many, like, tens of gigabytes of data just for, like, all the source code and, like, artifacts and stuff that I built, and I want that to come back next time, so.Matei Zaharia [00:16:51]: Yeah.Matei Zaharia [00:16:51]: But yeah.Matei Zaharia [00:16:52]: Before the show, we was talking about some statistics that might be surprising at the adoption.Matei Zaharia [00:16:56]: It could be internal, it could be external, whatever comes to mind, just to impress people the scale this is happening.Swyx [00:17:02]: So we, on the analytics side, I think we launchedReynold Xin [00:17:06]: Maybe 50 or 60 million virtual machines a day across all three clouds, so we're one of the biggest compute orchestrators out there.Reynold Xin [00:17:13]: Stuff for sure for CPU compute.Swyx [00:17:14]: Yeah.Matei Zaharia [00:17:14]: Yeah.Reynold Xin [00:17:15]: the. And all of this process, I think exabytes of data, I joked about depending on which time zone you are, typically before you have breakfast, Databricks would have processed exabytes of data already on that day. and on Neon, it's pretty interesting, too. It's launching, I think, 13 million databasesSwyx [00:17:34]: YeahReynold Xin [00:17:34]: a day now.Swyx [00:17:35]: Yeah, to me that was, like, aReynold Xin [00:17:36]: And that's just likeSwyx [00:17:37]: Like, what do you mean?Matei Zaharia [00:17:38]: Yeah. And that's the point.Reynold Xin [00:17:40]: And a lot of those were thanks to agent- agents and branching experimentationSwyx [00:17:44]: YeahReynold Xin [00:17:44]: because we made it so easy and so quickly, and thanks a lot to Nikita's team, to launch databases. It's, the. So it's changing the way people use databases.Swyx [00:17:54]: Yeah. Okay, we're gonna go into more database talk in a bit, but I wanna make sure we close up anything on Omnigentt. you mentioned, you were excited about the securityOmnigent Security, Contextual Policies, and Spend ControlsSwyx [00:18:03]: control side.Matei Zaharia [00:18:04]: Yeah.Swyx [00:18:04]: a lot of companies are figuring that out right now, as well as the spend side.Matei Zaharia [00:18:08]: Yep.Swyx [00:18:09]: what have you found there?Matei Zaharia [00:18:11]: Yeah, so I spent quite a bit of time talking to internal users, developers, security team, managers, and also lots of customers, and there's a few things. Like, first of all, one thing, that immediately was. became obvious is for security, there's this tension between, like, usability and security. And, the way people do. Like, a lot of coding agents today have very basic things like you can tell me which tool patterns I'll allow or disallow or whatever. It's like yes or no. But that puts you in a very tough spot. So just as an example, like, should my agent be able to read, some confidential documents, or let's say, should it be able to install new packages from npm, which, maybe it's compromised. Yes or no? Like, maybe I wanna allow it. Should my agent be able to publish stuff to the company website? Well, if I'm using it to code on the website, yes. But should it be able to do both, so it can, like grab a confidential document and be prompt injected and leak it? Probably not. So the thing we decided we need is stateful or what we call contextual policies where you keep track of the state of that session. It's not like is it allowed to push to the marketing site or not, but, like, hey, if it did a risky thing, like it installed, a old package from npm, or it read, like, 1,000 confidential docs, then no. Then don't, don't do it. Otherwise, maybe it's okay. That's one example of, like, moving that trade-off so it's both more secure and more useful by having a more powerful engine, essentially. This requires tracking sessions. The other piece that was interesting there is, like, there are these very level events it's doing, and you want some libraries on top that parse them. Like, for example, we have a, MCP server on Google Drive internally. It's got 60 API calls. like, how do I know which of those, like, will share a document with stuff on the internet and which ones won't? It's, it's annoying. So we designed in Omnigentt the policy layer so that it's functions and you can have libraries. Like, someone can make something that maps the level events to high-level ones, and then you write a policy about the high-level things that came out. so and thatSwyx [00:20:25]: This is related to the Panther,Matei Zaharia [00:20:27]: Yeah, Panther is. will help with that. PantherSwyx [00:20:30]: YeahMatei Zaharia [00:20:30]: a similar idea on the event processing side, and it's Python-based versus a weird custom language. this is more, as in realSwyx [00:20:39]: I didn't even know we were good yeah.Matei Zaharia [00:20:41]: Those things are happening, yeah.Swyx [00:20:42]: Yeah.Matei Zaharia [00:20:42]: So yeah, but these are the cool things. I think the contextual or stateful part, and then the way it can be libraries, and that was another reason to make it open source because others will write libraries and, like, we and our customers can use them. And the final thing, because it's stateful, one of the states we track is how much you spent in that session. So I can. I've had, like, I ask an agent to debug something, and it spent $500 because it decided to read a lot of log files and burn a lot of tokens. but I can literally say, “Okay, launch a agent to do this and cap it to spending $5.” Like, ask me for permission if it needs more. And because we're counting that within that session, it'll pop up and tell me, “Okay, you spent five, $5. Do you wanna go on?”Reynold Xin [00:21:27]: So important context here. Matei spent the last five years, a lot of his time was architecting Unity Catalog at DatabricksMatei Zaharia [00:21:34]: YeahReynold Xin [00:21:34]: which is the governance layer for data.Matei Zaharia [00:21:35]: That's right, yeah.Reynold Xin [00:21:36]: And he's combining expertise at that layer together with all the AI governance he knows.Matei Zaharia [00:21:41]: Yeah.Swyx [00:21:41]: DoMatei Zaharia [00:21:41]: But I also spent a lot of time being annoyed by coding agents and getting prompts.Matei Zaharia [00:21:46]: And also as theReynold Xin [00:21:48]: All the aboveMatei Zaharia [00:21:48]: I don't want to end up on the front page as, like, I installed some weird npm package and leakedSwyx [00:21:53]: YeahMatei Zaharia [00:21:53]: all the code, so I'm especially paranoid. But also I have very little time, so I don't want to sit there approving, like, do you want to run a 20-line, bash script, yes or no? so that's why I spend a lot of time figuring out, like, how can I make it as safe as possible and not annoying?Swyx [00:22:10]: Yeah. Is safety and mmm, let's call it security a bigger concern than token maxing or token budgets? which one is, likeMatei Zaharia [00:22:19]: Oh, yeah, they're both there. I don't know. I guess it depends on the type of company you are. So I think, some companies, like, the budget is, limited and, they really care about thatSwyx [00:22:34]: you can be Uber and still be concerned?Matei Zaharia [00:22:36]: Yeah. Oh, yeah, totally. Yeah. If you haveReynold Xin [00:22:38]: for us, securityMatei Zaharia [00:22:39]: YeahReynold Xin [00:22:40]: super paramount.Matei Zaharia [00:22:40]: For us, security is absolutely critical as a, cloud provider. It's, it's the most important thing, and, token maxing, we're not so worried about it yet, but I've seen the Like, for example, I talked to some consulting companies. They have, like, 100,000 employees who are all coding for customers. If those each spend, like, an extra $1,000 a month, that's, that's not fun.Swyx [00:23:04]: YeahMatei Zaharia [00:23:04]: we have, like, only a few thousand engineers.Swyx [00:23:06]: What's the policy in Databricks? Is it just unlimited or what'Matei Zaharia [00:23:08]: It's, it's unlimited, but we do. we use our own product to, like, analyze the traces and stuff, and we have a team that'looking to optimize and to see if anyone's doing something weird. And, we had some really cool insights just from analyzing current traces, like whichSwyx [00:23:24]: YeahMatei Zaharia [00:23:25]: models are better at, say, Rust versus like TypeScript or whatever. So yeah, at least in our code base.Swyx [00:23:31]: Yeah. Amazing. Obviously, I have to ask the token question, obviously.Matei Zaharia [00:23:34]: Yeah.Swyx [00:23:34]: I think it'sReynold Xin [00:23:34]: YeahSwyx [00:23:34]: it's a key thing. But yes, security and control above that, and figuring out a sane layer there you can have some autonomy, but, not too much.Matei Zaharia [00:23:43]: Yeah. Yeah, and we wanna make it super easy. As a engineer, you should set a thing. So in Omnigentt, you can ask your agent, “Set a policy on yourself to do this.” So it can likeSwyx [00:23:52]: But if there's something I should be showingMatei Zaharia [00:23:53]: YeahSwyx [00:23:53]: I don't, I don't see it on the GitHub, but,Matei Zaharia [00:23:55]: Oh, yeahSwyx [00:23:56]: there's justMatei Zaharia [00:23:56]: Well, in the docs there's something.Swyx [00:23:57]: Yeah, this is it.Matei Zaharia [00:23:58]: You can look at it later.Swyx [00:23:59]: Okay. Yeah.Matei Zaharia [00:23:59]: Just look in the docsSwyx [00:24:00]: YeahMatei Zaharia [00:24:00]: contextual policies if you wanna see.Swyx [00:24:04]: I just like to point peopleMatei Zaharia [00:24:05]: look at the built-in policies.Swyx [00:24:06]: Yeah.Reynold Xin [00:24:06]: Yeah.Swyx [00:24:06]: If you want to, follow up on this is exactly where to look, right?Reynold Xin [00:24:10]: Yeah.Matei Zaharia [00:24:10]: Yeah. yeah, and the story of these is, like, I just wrote, like, I wrote a doc with like 10 ideas for things before as you were working on them. Well, that was, like, my wish list of things people asked, and I told the team, like, “Hey, can you do like at least five of these for the launch?” And then they just got back with all of them, so.Swyx [00:24:29]: Oh, wow.Matei Zaharia [00:24:29]: so you can come up with more, but them- some of them are just meant to be examples. really you can intercept, like, any event the agent is making, and you can then either block or force it to ask the user or, like, allow, and you can update state to keepSwyx [00:24:45]: YeahMatei Zaharia [00:24:45]: track stuff.Swyx [00:24:46]: Yeah, ‘cause ultimately you're, I think of you as, like, a systems designer.Swyx [00:24:50]: You let people plug in, right? That's the wholeMatei Zaharia [00:24:51]: YeahSwyx [00:24:52]: modus operandi of what you do.Matei Zaharia [00:24:53]: Yeah.Swyx [00:24:54]: It's likeMatei Zaharia [00:24:54]: And we care a lot about also composab- like, can someone else write a library that others use, whichSwyx [00:24:59]: YeahMatei Zaharia [00:24:59]: this is meant to.Reynold Xin [00:25:00]: There's also a batteries included philosophy hereMatei Zaharia [00:25:03]: YesReynold Xin [00:25:03]: probably very similar to how you did Spark, which is you could just start using.Swyx [00:25:06]: Yeah.Matei Zaharia [00:25:06]: Yeah, that's right. It has to be good out of the box at certain things, and then you can build your own things on top that, like, we don't wanna do. But in Spark, if you just wanna like, I don't know, like read a table or do, like, a aggregation, it should be awesome at that out of the box.Building on Omnigent: Contributions, Startups, and AnalyticsSwyx [00:25:23]: Yeah. People wanna catch up on Omnigentt, they should watch your keynote.Swyx [00:25:26]: they should go through the GitHub and the docs. If they wanted to contribute, or they want to build on this ecosystem what would you call out as the most high-leverage places get involved?Matei Zaharia [00:25:36]: Yeah, do get involved in the Discord and in GitHub. Our team is there, is monitoring, and, some of the things people ask for we just built ourselves. Some of them, we're, we're collaborating with them to build it. and also tell us, likeSwyx [00:25:49]: Yeah, they're gonna be veryMatei Zaharia [00:25:49]: how you would like to use it because I think especially for developers, like, everyone wants it to work their own way, and a really good developer tool, like you have to hear the feedback on all the ways and figure out the abstractions and how to let people customize. So we'd love to hear, like, if you think, “Hey, I, I don't want it to work this way,” tell us. We really just wanna get that compatibility layer across agents and then let you do stuff on top.Swyx [00:26:14]: Yeah. is there any, in terms of like the startup side, I'm, I'm a founder.Swyx [00:26:18]: I wantMatei Zaharia [00:26:18]: YeahSwyx [00:26:18]: I see an opportunity, I wanna get in front of you. What's your request for, like, a startup that, like, I wish someoneMatei Zaharia [00:26:23]: Oh, like you wanna integrate with us?Swyx [00:26:24]: someone was working on this.Matei Zaharia [00:26:26]: Oh, for a startup?Swyx [00:26:27]: Yeah.Swyx [00:26:28]: Like, your, you got your own startup. It's doing well.Matei Zaharia [00:26:30]: Yeah.Swyx [00:26:30]: But like, if you weren't working on your own startup, what is, like, obvious that you should You advise many startups too, obviously.Matei Zaharia [00:26:37]: I do think, just as a company with a lot of engineers, like anything that helps me make sense of how people are usingSwyx [00:26:46]: SpendMatei Zaharia [00:26:46]: coding agents and,Swyx [00:26:48]: Yeah. AnalyticsMatei Zaharia [00:26:48]: spend, but also quality or like you should write, you should add this skill, or you should write this thing, or your agents are really horrible at tasks involving this service, so I go spend time. That would be nice. yeah.Swyx [00:27:00]: Yeah. The closest I've found is, this team, GitAI.Matei Zaharia [00:27:03]: Oh, cool. Yeah.Swyx [00:27:04]: They started with, like, we will just do, code and human attribution, but they're building the analytics layer on top of that.Matei Zaharia [00:27:12]: Yeah.Swyx [00:27:12]: I do think, like, there are a bunch of, like, artificial analysis is obviously,Matei Zaharia [00:27:18]: Yeah, they have their benchmarksSwyx [00:27:18]: doing super wellMatei Zaharia [00:27:19]: YeahSwyx [00:27:19]: with their stuff. so there's, there will be people. I think this is like the domain of consultants first, but then peopleMatei Zaharia [00:27:26]: YeahSwyx [00:27:26]: will build software that, let's say, it's kinda like the management planeMatei Zaharia [00:27:29]: YeahSwyx [00:27:30]: for coding agents.Matei Zaharia [00:27:30]: Yeah, I think there'll be a lot of insights there. You have it in other areas.Swyx [00:27:34]: Okay. Well, and then the other, big thing is your dream engine.LTAP: Lake Transactional/Analytical ProcessingSwyx [00:27:39]: maybe you wanna tell the story of, LTAP.Reynold Xin [00:27:45]: So, and background with. I'm, I'm gonna make people listen to our Ankur Goyal episode where we talked about SingleStore, HTAPMatei Zaharia [00:27:52]: YeahReynold Xin [00:27:52]: and all that history.Matei Zaharia [00:27:52]: Yeah. The LTAP idea is pretty simple. so if people have heard of the, Ankur's, talk about HTAP, it's effectively the world of databases. Sorry, there's like maybe a lot of context needs to be injected here. The world of databasesSwyx [00:28:06]: I am happy to be the database podcast that I'm forcing people to, like, learn your databases, guys.Swyx [00:28:11]: You cannot vibe code with just markdown files.Reynold Xin [00:28:13]: Yeah.Swyx [00:28:13]: Like,Reynold Xin [00:28:14]: It's one of the most important fundamental systems technologies out there. But the world of database effectively split into roughly two halves. There's what we call OLTP databases, which are transactional, and think of your Postgres, your MySQL, your Oracle databases, and the other side is what we call analytics, and sometime might refer to term OLAP. And the difference is on OLTP, you typically have maybe run some transaction on some event that looks up at one specific row. We update that row, right? It's a very oriented data structure. And on analytics, you're trying to reason on the data. You're trying to compute, “Hey, what's my revenue per store? What's my. How's my website doing every day?” And then you, eventually want to probably end up running anal- machine learning on it to predict, “Hey, how will my maybe sales be going in the future?” they are so very different architecture, and everybody start with OLTP databases. Every app, when you become serious enough, that needs more than markdown files, you need to have a database. You want to lose your data, you want to have some transactional consistency. But once you want to reason on the data, if you only have like- A hundred rows, it's probably okay to run it on your Postgres or your own, your MySQL database. But once you have more data and want to run more complicated analysis, the very analysis might crush your Postgres database. So you start doing, getting data out of the OLTP databaseSwyx [00:29:35]: Replication.Reynold Xin [00:29:36]: Replicate them into the analytic systems and just startSwyx [00:29:39]: Yeah, which for people, Elasticsearch is, like, aReynold Xin [00:29:42]: Yeah. So some of them get into Elasticsearch for, like, blocked analysis. A lot of our customers obviously get into Databricks to run more sophisticated things.Swyx [00:29:51]: Yeah.Reynold Xin [00:29:51]: And there's this term called CDC, whichMatei Zaharia [00:29:54]: Change data captureReynold Xin [00:29:55]: change data capture. and what it does, it reads the binlog of the database, and if you don't understand what binlog is, it's fine. The, but it's a little delta of the data, and it reconstructs based on the delta, the state of the database, on the analytics side. But CDC is, like, a very painful thing. It's how standard in the industry, everybody uses it, but, it ends up being. I think many data engineers ends up being waken up at, like, 3:00 a.m, because there's some pipeline thing.Swyx [00:30:22]: my explanation is, like, Airbyte is like a, became a $5 billion company just doing CDC.Reynold Xin [00:30:27]: Yeah, exactly.Reynold Xin [00:30:28]: CDC is, like, a veryMatei Zaharia [00:30:30]: It's hard.Reynold Xin [00:30:30]: It's one of the most boring but one of the most fundamental operations, like, powering modern society.Matei Zaharia [00:30:37]: huh.Reynold Xin [00:30:37]: But it's so brittle that, we joke that it's, should be called continuous data corruption, because you might change your schema on your OLTP database, and then the CDC pipeline fails to handleSwyx [00:30:48]: YeahReynold Xin [00:30:48]: the schema change.Swyx [00:30:49]: Yeah.Reynold Xin [00:30:49]: And then everything goes out.Swyx [00:30:51]: And there's all sorts of tricks that you can do, like, you add in, like, some versioning or whatever, but yeah.Reynold Xin [00:30:55]: Yeah, but it's a very, in general, very complicated. Like, I think at my keynote, I asked the audience put up their hand if they love their CDC pipeline. Only, like, maybe two people put it up. So if single store, like, about maybe a decade ago, I think the industry had this idea, hey, what if I built a single database that can handle both workloads? Now I don't.Swyx [00:31:12]: Which, like, by the way, every database person ever has ever always dreamed about this.Reynold Xin [00:31:15]: Yes. Yes.Reynold Xin [00:31:16]: This is the holy grail of database engineering is why not build a single system that can do both of this? But it ends up just being a lot of compromises. one, I think one of the first issue is that, hey, each. they say Postgres has a massive ecosystem, right? You want to be using the tools that's built for Postgres. And Spark, for example, had a massive ecosystem. There's a lot of libraries you want to use. If you were to create now a new thing, you don't have a ecosystem. You tend to create a new, smaller proprietary API, and you're lacking both, and it's also very difficult to make it performance-wise to be, comparable on either side. So it ends up being sucking on both. And our whole idea of LTAP, it's obviously a wordplay on the term HTAP, is that we think this is HTAP done right. HTAP wants to build a single engine for both. We think you can get 99% of what you need by unifying the storage, and just have a single storage layer. And once you have the single storage layer, if your Postgres databases are writing data in a column-oriented format, everything analytics can just go read that data directly without any delay, right? There's no pipeline in between, so all the data will immediately be available for reasoning analytics. I think I was telling some customers earlier, hey, when we talked about this is gonna be super useful for agents, I at first didn't really believe in it myself, even though we wrote that positioning.Lakebase, Agents, and Live Operational DataMatei Zaharia [00:32:39]: Yeah.Reynold Xin [00:32:40]: But then last night I was having dinner with a Australian customer, and they told me, “Oh, hey, one of the big issue we have is we have all these logs from our services, and we see SLA dips and want to investigate. But then there's no way for those agents to even understand what's going on in the actual databases themselves. All we see is just, like, product telemetry of the database and the services.” It would make those agents 10 times more powerful if understand, for example, who's placing those orders, what is happening, what exactly are they doing. So now I'm sold on our own message.Swyx [00:33:13]: Yeah.Reynold Xin [00:33:14]: I think it's really. It gets you the almost all of the benefits of the HTAP holy grail, which is, hey, make the data available immediately for reasoning analyticsSwyx [00:33:26]: Yeah, I think,Reynold Xin [00:33:27]: without compromiseSwyx [00:33:28]: in the way that humans are generally intelligent and want to have the ability and access to query anythingReynold Xin [00:33:34]: YeahSwyx [00:33:35]: while they do the work, they also need history and need context.Swyx [00:33:38]: And, like, where else does they get context? That's it's an analytical workload.Reynold Xin [00:33:41]: Exactly.Matei Zaharia [00:33:42]: Yeah. Yeah. And I remember when we had incidents with our databases and engineers said, “Well, I can't just run a giant query on it to see what's going on because that's gonna bring down the database and hoard it even more.” Like, that's the stuff that this gets rid of, because you spin up a whole separate fleet of machines that's doing the analytics. You're not overloading, like, the main databaseReynold Xin [00:34:02]: RightMatei Zaharia [00:34:02]: that's still trying to serve stuff.Reynold Xin [00:34:04]: Yeah.Matei Zaharia [00:34:04]: Yeah.Why LTAP Works Now: Parquet, Postgres, and LakebaseSwyx [00:34:05]: So this has been a dream for a while. what had to get done in order to get to today? Like,Reynold Xin [00:34:11]: Yeah.Swyx [00:34:11]: I feel like, you have announced variants of this several times, but it wasn't as clear as LTAP.Reynold Xin [00:34:18]: Yeah.Swyx [00:34:18]: I think LTAP is like Like, okay, we've got it, guys.Matei Zaharia [00:34:21]: This thing, yeah.Reynold Xin [00:34:21]: I was talking to somebody at Meta, and then he was asking me, “Hey, what's the catch? Why is it possible now?” And I think the reality is we took a lot of time to work on the Lakebase architecture. obviously a lot of it came from the Neon team, which is a separation of storage from compute. And it turned out it was just a tiny little step away going from that to this LTAP idea, which is, hey, we just. in the Neon architecture and in Lakebase architecture, we're writing data in oriented format to the open data lake, but in there we're writing in Postgres pages. Ali and I were spending a lot of time debating, hey, can we just change that to write in column-oriented format? And we're just debating, and one day, one of our engineers who's, like, super smart came in, he's like, “Hey, I just prototyped it. It works.”Swyx [00:35:07]: Wait, it's, prototype what?Reynold Xin [00:35:09]: Prototype, instead of storing the data in the data lake in the oriented formatSwyx [00:35:15]: ColumnReynold Xin [00:35:15]: like Postgres pagesSwyx [00:35:15]: YeahReynold Xin [00:35:16]: write them in Parquet.Swyx [00:35:17]: Yeah.Reynold Xin [00:35:18]: and he just made the observation that, hey, our storage fleet has a lot of extra idle CPUs And we could use those CPUs to do the transcoding from row to column, where row is good for OLTP, but column is good for analytics. so let's do that transcoding at that time. And as a matter of fact, once you transcode the data compresses better. So from those services writing to, for example, S3 or other data lake, like object stores, you can write them faster ‘cause now they are now smaller.Matei Zaharia [00:35:49]: Yeah.Reynold Xin [00:35:49]: So there's no overhead, it's no compromise in performanceMatei Zaharia [00:35:52]: Some CPU overhead.Swyx [00:35:54]: Yeah, because,Matei Zaharia [00:35:55]: YeahSwyx [00:35:55]: we had extra CPUs anyway.Matei Zaharia [00:35:56]: We had that fleet anyway, yeah.Swyx [00:35:57]: so the debate ended. it's one of the classics of, tech, issue of a lot of debate, but then somebody went ahead and just tried to prototype it and it worked.Matei Zaharia [00:36:06]: But, like, something this strategicSwyx [00:36:07]: That's rightMatei Zaharia [00:36:07]: and important to the company, I expect there to be, like, a kickoff thing, like a design doc. Nothing like that.Swyx [00:36:13]: Nothing like that.Swyx [00:36:14]: He just. We were debating in many meetingsMatei Zaharia [00:36:17]: Yeah.Swyx [00:36:17]: and then we're just debating whether it's possible or not from first principle.Matei Zaharia [00:36:20]: YeahSwyx [00:36:20]: and then, somebody just did it.Matei Zaharia [00:36:23]: Yeah, if you set yourself up so people do that'll be great. And that happened a bit with Omnigentt too. I think if I just had a doc on, like, we can make these together, everyone would, would think, “Oh, what about this? What about this?” But then you. if you try it out, it helps. And then if you have real users and they bash it and, like, it's still working, or in this case, if you have the workload, what the workload looks like, you can just test the same pattern then.Databricks' Culture of Fast PrototypingSwyx [00:36:47]: Yeah.Matei Zaharia [00:36:47]: Yeah.Swyx [00:36:47]: Tech aside, which is very cool, this is, like, the most important thing, the culture of innovation, and you don't have to ask my permission, you don't have like, do a whole form- formal process, just do it?Matei Zaharia [00:36:59]: Well, especially these days, I think withSwyx [00:37:01]: YeahMatei Zaharia [00:37:01]: AI, it's easier to buildSwyx [00:37:02]: But so, likeMatei Zaharia [00:37:03]: a prototypeSwyx [00:37:03]: I think you are very I made a lot of suite of, like, large companies and, like, I think that at scale, things slow down, and I'm sure you felt it already, but somehow you have this core of people that, like, are exempt. How? I think we hire and we work with really good people, and that's a very important part of it, and empowering them, but also spending a lot of time, maybe us in the trenches matter a lot also.Matei Zaharia [00:37:28]: Yeah, I think, I think first, people can adapt to being in the larger company, so that helps. And we wanna make sure they know that they can try stuff and settle debates and have a lot of examples of how it was done before, or launch a thing in beta or whatever. and then the other thing I do think as a company, like despite the size, we don't launch that many, like, products. We try to keep it pretty coherent. That's, that was the whole, like, theory of the company, was like instead of having, like, 20 Amazon services you need to set up, like a analytics and machine learning stack, you just have one, and it's, like, the same API, the same semantics across all of them, the same copy of the data. So that requires, like, unification. And then we added one more thing at a time. Like, we added storage with Delta Lake. We didn't used to do any storage. Then we added SQL, we added, machine learning platform stuff. So, but yeah, don't, don't do too many, but do those things well and, that also helps, it helps keep it manageable.Reynold Xin [00:38:33]: Yeah. The other thing we encourage a lot is instead of building, boil the ocean for everything, let's figure out how do we do it incrementally, how do we do it very quickly. Like, many of our productsMatei Zaharia [00:38:43]: YeahReynold Xin [00:38:43]: they're built in the span of weeks, and then we go to, hey. Like, usually my first question to whoever team is building is who's the target customer? Who are you working with? Are you on a first-name basis with them? Are you texting with them? I think having that very tight loop,Matei Zaharia [00:38:59]: Can you bring up another launch that comes to mind when, in this thing? I just want to give examples.Reynold Xin [00:39:04]: Omnigentt itself happened that way.Reynold Xin [00:39:05]: Yeah.Matei Zaharia [00:39:06]: Who's the customer? That's a good oneReynold Xin [00:39:34]: storage layer we did. we had, our largest customer at the time said like, “Okay, I need some. I want something in the cloud ‘cause, I. if the rest of our network is compromised, like this thing needs to be separate to store and query the events.” And then, talked to us, he said, “Okay, this is the rate of events per second. This is, like, the freshness I want. Can you do it?” So that was, like, way larger than any workload we had, and we had our, engineer, working on that, Michael Armbrust, and he worked just to make this work. And once it worked for them, it worked for everyone else. Yeah. This was early in the company, probably like four years in or something.Matei Zaharia [00:40:24]: 20- 2018?Swyx [00:40:26]: Yeah, ‘17, ‘18.Matei Zaharia [00:40:28]: Few companiesSwyx [00:40:28]: Do you have other examples?Matei Zaharia [00:40:30]: there'Swyx [00:40:31]: Maybe you have othersMatei Zaharia [00:40:31]: yeah, Clean Room, which is how you share data in a way without sharingSwyx [00:40:35]: YeahMatei Zaharia [00:40:35]: underlying data, but you allow specific operations. Those were done effectively initially just for two customers. I think the industry has a sense of, hey, maybe if you overfit to, like, one or two customers, it's gonna be really bad for you. But I think the, downside of overfitting is much smaller than the upside itself. And if you try to be too ambitious and boil the ocean, it's a much bigger problem.Swyx [00:40:58]: Yeah. Yeah.Matei Zaharia [00:40:58]: ‘Cause you might end up having no customer.Swyx [00:41:00]: Yeah, that's more, that's the more likely outcome.Matei Zaharia [00:41:02]: Yeah.Tech Companies vs. EnterprisesSwyx [00:41:03]: than you can pivot from there. I do think there is such a thing as a bad customer that sometimes you should fire. Yeah.Matei Zaharia [00:41:08]: They could exist sometimes if you drive. well, one of the challenge I think we probably see, and maybe many AI, so newer generation companies are seeing is, so tech companies are very different from tech companies or traditional enterprises.Swyx [00:41:22]: Yeah.Matei Zaharia [00:41:22]: And, if you optimize everything just for tech companies, you might have various challengesSwyx [00:41:27]: OhMatei Zaharia [00:41:27]: scaling them outside of tech companies.Swyx [00:41:28]: Okay, what likeMatei Zaharia [00:41:30]: YeahSwyx [00:41:30]: what like top three differences that you always think about?Reynold Xin [00:41:33]: Governance is a big oneMatei Zaharia [00:41:34]: I think, yeah, a big one is like, yeah, security, data privacy, governance, all that stuff. So usually if you're building some kinda like B2B or developer tool, like your biggest market is gonna be enterprises, but it's just very different. A company that's existed for like, it's had some form of IT for like 30 years, they have so many legacy systems or they operate in a regulated space. whereas a startup or, even like a, like sorta more recent tech company, all the. everything is new and pristine. So yeah, it's just different, and if you've never worked with enterprises or been in one, you just won't know about it.Reynold Xin [00:42:13]: Yeah.Matei Zaharia [00:42:13]: Yeah.Reynold Xin [00:42:13]: And the procurement process is probably quite different. There's far more stakeholders.Matei Zaharia [00:42:17]: Yeah, that is one. Yeah.Matei Zaharia [00:42:18]: Another piece that's interesting is I think some tech companies, people, will say, “Oh, I can build that myself,” right? I'll just build that myself.Matei Zaharia [00:42:27]: So then you go,Reynold Xin [00:42:28]: I don't think people say that about Databricks, butMatei Zaharia [00:42:31]: yeah, it dependsReynold Xin [00:42:32]: They do.Matei Zaharia [00:42:32]: They do?Matei Zaharia [00:42:32]: Yeah, the. Yeah, and it depends on the teams and things. So, but, on the other hand, like many of the enterprises say, “I don't, I never wanna be in the business of building that.” Like, I don't want my, whatever, I'm a retailer or something, I never wannaReynold Xin [00:42:45]: Yeah, sell clothes,Matei Zaharia [00:42:46]: be down because like some weird like nerd like couldn't get streaming pipelines working.Matei Zaharia [00:42:51]: That is not what I'm doing.Reynold Xin [00:42:53]: Yeah.Reynold Xin [00:42:53]: Yeah. This makes them great customers, to be honest, right?Matei Zaharia [00:42:55]: Yeah. But you have to understand that it's hard without having worked there and stuff, like you may not appreciate.Reynold Xin [00:43:01]: Look, I think they're all great. don't get me wrong, they have different challenges. But the, many of the tech companies, for sure there's a lot, far more DIY.Matei Zaharia [00:43:10]: On the flip side, you have people who are. they're very much experts in their domain, like they're building airplanes, they're, designing medicines, whatever, and they just want to bridge the technology, where like they don't wanna learn, databases or whatever. As cool as we think it is, even as interesting as the average software engineer might think it is to read a little bit, like they just never wanna know. They just say, “I have a, giant like, matrix or whatever with my, clinical data, like how do I, how do I like cluster it or whatever?” So yeah.The Dream Engine and Rewriting the Database StackReynold Xin [00:43:40]: Yeah. That's true. Okay, so and then I wanted to build out the dream engine, vision. where does this all lead? So one of the thing we, realized maybe a couple years back is that every single database engine out there, especially on the analytics side, are a decade old. pretty much everything that have reasonable traction are about a decade old. And they all started targeting some very specific narrow use cases, and then over time it's become more and more successful. They have grown in their ambition, and then they try to support more and more use cases. But the fastest way to support those use cases tend to be hacked around the abstractions that were initially created, that were not for those use cases.Matei Zaharia [00:44:23]: Yeah.Reynold Xin [00:44:23]: And then, but you can support them more or less okay. And before it, after 10 years of organic evolution that way, it becomes a gigantic pile of s**t.Reynold Xin [00:44:31]: the. And, but that includes Databricks. And very few company or very few systems, I think, have the gut to say, let's go start from scratch. Let's go back to the drawing board and design, knowing everything we know today after a decade of workloads and probably billions in revenue, let's attempt to rewrite it from scratch and make sure it will work and it can support all of these use cases. So we started doing that, but it's a very ambitious project. by the way, you can search on Wikipedia, there's this thing called second system syndrome.Matei Zaharia [00:45:08]: Yeah, I know that. Yes.Reynold Xin [00:45:09]: Or second system effect.Matei Zaharia [00:45:11]: Every developer must know what a second syndrome is.Reynold Xin [00:45:12]: It's you built your first thing and it works out great, and the second one's bound to fail because you become too ambitious.Reynold Xin [00:45:19]: And then you ask so many requirements.Matei Zaharia [00:45:20]: Or like you think everythingReynold Xin [00:45:21]: YeahMatei Zaharia [00:45:21]: and then you're likeReynold Xin [00:45:22]: You justMatei Zaharia [00:45:22]: you're, “I'm gonna design the perfect system this time.”Reynold Xin [00:45:24]: Yeah. And it turned out it's not perfect, and then it start failing and you're too ambitious, never launch, and you get killed. The, and the engineering team that started this, they were brilliant. I think we hired some of the best database engineers, on the planet into Databricks, and they were brilliant. Thank God it's not their second system. Many of them have built more than two in the past.Matei Zaharia [00:45:44]: Ah, nice.Reynold Xin [00:45:45]: But they were still worried about this, hey, building a database engine from scratch, I think the conventional wisdom is gonna take like five years to mature. This would be a very long-term project. It could fail. I think one of the engineers jokingly said, “Hey, maybe we just call it Reynolds Stream Engine.” If we name after a founder, maybe we then may get canceled or killed. But I think they built something pretty remarkable. they went back to. They changed the way the database engines were built from a paradigm point of view. Usually when y

In this episode we speak to Jonathan Grahl.  Jonathan is the Team Lead of Infrastructure at Lovable where he oversees the platform stack the company runs on. We talked about Kubernetes, Sandboxes and Chocolate.   Do you have something cool to share? Some questions? Let us know: - web: kubernetespodcast.com - mail: kubernetespodcast@google.com - twitter: @kubernetespod - bluesky: @kubernetespodcast.com   News of the week OpenTelemetry is a CNCF Graduated Project CNCF TAG Elections KubeCon India Kubernetes Community Days global events Links from the interview Lovable Cilium Etcd Cloudflare Sandbox Agent Sandbox (Kubernetes) OpenClaw Claude OpenAI Bun toolkit for Javascript gVisor Firecracker Kata Containers Vitess

Take the 2026 AI Engineering Survey and get >$2k in credits and AIE WF tickets!On the product side, everyone is getting Computer - Perplexity, Manus, Cursor, and so on. Meanwhile on the research side, agentic evals like TerminalBench and GDPVal are also assuming computer (Harbor). On both ends, the consolidating LLM OS stack has become a standard toolkit, and Daytona is one of a small set of AI Infra companies that are booming because of it.“The end of localhost” has been Ivan Burazin's obsession for more than a decade.Something that is all too familiar…Long before agents became the default way people talked about software development, Ivan was already chasing the idea that development should not depend on a fragile local machine. CodeAnywhere, one of the first browser-based IDEs, was an early attempt at that future: move the development environment into the cloud, make setup reproducible, and free developers from the endless “works on my machine” tax.The thesis was directionally right, but the market wasn't ready yet.However, agents changed that. They do not care about a laptop, desk setup, or favorite editor. They need a computer they can access through an API: something stateful enough to keep working, fast enough to spin up instantly, flexible enough to resize, isolated enough to be safe, and composable enough to run the messy real-world workflows that real software engineering actually requires.Daytona isn't just selling “sandboxes” in the narrow code-execution sense. It is the latest version of Ivan's original localhost thesis.In this episode, Daytona's CEO joins swyx to explain why AI agents need more than code execution boxes: they need composable computers, stateful sandboxes, instant startup, dynamic resources, and infrastructure that can survive workloads going from zero to 100,000 CPUs.We go deep on the new agent compute market: Daytona's hard pivot from human dev environments to AI sandboxes, the New Year's Eve MVP that customers begged for, why Daytona runs on bare metal with its own scheduler, how one customer runs almost 850,000 sandboxes a day, and why RL/eval workloads went from 0% to roughly 50% of usage in just months. Ivan also explains why agents need Windows and macOS machines, why CLI may matter more than MCP, why Kubernetes is painful for this workload, and why the future AI cloud may look more like Stripe than AWS.We discuss:* How Daytona grew out of CodeAnywhere, Shift, and the “end of localhost” thesis* Why Daytona pivoted from human dev environments to AI sandboxes* Why agents need composable computers instead of disposable code execution boxes* The New Year's Eve MVP that customers chased API keys for* Why Daytona chose bare metal, stateful snapshots, and its own scheduler* How Daytona spins up one sandbox in ~60ms and 50,000 sandboxes in ~75 seconds* Why Daytona's biggest customer runs ~850,000 sandboxes a day* How RL/eval workloads create zero-to-100,000 CPU spikes* Why RL workloads went from 0% to roughly 50% of Daytona usage* Why customers compare Daytona against EKS/GKS and say they're “never going back”* Why every AI agent may need a computer, including Windows and macOS environments* The Apple licensing constraints that make macOS sandboxes hard* Why CLI gives agents more power than MCP* How open source helps agents integrate Daytona* Why agent-generated PRs may break today's CI/CD assumptions* Why AI SaaS companies reselling tokens may face a cold shower* Why the AI cloud may look more like Stripe than AWSIvan Burazin* LinkedIn: https://www.linkedin.com/in/ivanburazin* X: https://x.com/ivanburazinDaytona* Website: https://www.daytona.io* X: https://x.com/daytonaioTimestamps* 00:00:00 Hook* 00:01:12 Introduction* 00:03:15 CodeAnywhere, Shift, and the end of localhost* 00:05:58 What Daytona is: composable computers for AI agents* 00:08:07 The pivot from dev environments to AI sandboxes* 00:10:17 The New Year's Eve MVP and customers begging for API keys* 00:12:56 Bare metal, stateful sandboxes, and Daytona's scheduler* 00:17:28 60ms startup, 50,000 sandboxes, and 850K daily runs* 00:21:53 Spiky RL/eval workloads and the new agent infra problem* 00:28:12 RL workloads, Kubernetes pain, and dynamic resizing* 00:33:31 Why every AI agent needs a computer* 00:38:48 macOS sandboxes and Apple's licensing problem* 00:44:28 Why CLI may matter more than MCP* 00:48:11 Open source, GitHub stars, and agent integration* 00:53:11 Git, CI/CD, and agent collaboration bottlenecks* 00:58:15 Founder life and building a 25-person infra company* 01:02:44 AI SaaS, token resale, and API-first business models* 01:06:10 GPU sandboxes, data centers, and compute growth* 01:09:48 Why the AI cloud may look more like Stripe than AWS* 01:11:26 Closing thoughtsTranscriptIntroduction: Daytona, CodeAnywhere, and the End of LocalhostSwyx [00:00:02]: Okay, we're in the studio with Ivan Burazin, CEO of Daytona. Welcome.Ivan [00:00:07]: Thanks for having me, man.Swyx [00:00:08]: Ivan, you and I go back.Ivan [00:00:10]: Way back.Swyx [00:00:11]: How I don't even know how, you found, did you reach out or, for Shift.Ivan [00:00:17]: I reached out to you. The reason was you - we were just - we were thinking about I was one of the co-founders of CodeAnywhere, the first browser-based IDE, and so we were thinking a long time of, localhost should die. And you had this article.Swyx [00:00:29]: End of localhost.Ivan [00:00:30]: Then I reached out to you because of that, and then we talked, and I was actually at a different job and learning about I was the head of, developer experience, and you were quite well-versed in that, and I actually reached out to you, among other people, how do we go about that? What are the key things and whatnot at this point in time? And you were nice enough to take the call, and I remember I was late on your call with you.Swyx [00:00:51]: I don't remember.Ivan [00:00:52]: I remember because I was with my then I'm thinking of a girlfriend or wife at that point in time, I'm not sure. It's the same person, so that's great, and I was late ‘cause we were, in, Italy on, vacation, and then I was late for something. I felt so bad, and you were so nice to be, good about.Swyx [00:01:10]: The reason I'm nice is because I'm also late to other people, so it's like, who's, who's without sin here, yeah, so I have to, for those who don't know, InfoBip Shift, there's this whole thing that, you did in the past, and, and that was basically one of the inspirations for me starting AI Engineer, which is like, I have to thank you for giving me that push to be like, “Oh, you can, you can build and sell conferences?”Ivan [00:01:34]: I remember you asked you asked me at the beginning to give me advisory shares, and I was so focused on what we were doing, I said no, and I should've took the advisory shares. So I'm sorry, dude. But anyway.Swyx [00:01:43]: We're not, we're not venture backed.Ivan [00:01:44]: No, it doesn't matter.Swyx [00:01:45]: It's Yeah, anyway, so I think what's impressive about you is that CodeAnywhere is the thing that you've been trying to build, and, you kind of put it on hold and then came back after InfoBip. Just give us the story, do you - the story and the origin story, going into Daytona.From CodeAnywhere and Shift to DaytonaIvan [00:02:05]: Sure. Like, really way back, me and my co-founder have been together. I say this, I've said this multiple times, it's like we were married and divorced and married. Some people actually ask me is my co-founder my partner. they thought it literally. It's not literally, but we have done multiple companies together, and to your point, we had this shift where we went from the CodeAnywhere to the conference called Shift, and then back to, Daytona. We originally started stacking servers, doing like virtualization in the early 2000s and, routers and doing basically all these things, at a foundational level, and that was a services company which we sold to focus on what my co-founder actually invented, which was the very first browser-based IDE, right, I say the first. Before us was actually Heroku. They did it for a very short time until they became Heroku. But outside of them, we were the only one, and it was called.Swyx [00:02:55]: There was Cloud9.Ivan [00:02:57]: Cloud9 came out slightly after us. There was Replit, which came out when we stopped doing it, Replit came out, and they have been successful since then, which is great. There was Nitrous.io. There was quite a few that existed at the time, but it was like too early. But the interesting part is that we, at that point in time, because there was no VS Code, there was no Kubernetes, and Docker had just started when we Or I'm not sure if it was even public at that point in time. And so we had to build everything to the whole stack ourselves and that was the key learning that we brought into and that we've been using in Daytona today. So it was super early. There's about 3 million people used CodeAnywhere. It was slightly, it was angel-backed more than venture-backed. We ended up paying everyone back because it didn't have that sort of scale. But, three years ago, we started something similar with Daytona, which is not what we are today, but it was automating dev environments for human engineers, the basically the underlying stack of CodeAnywhere. And then we did a hard pivot last January to sandboxes. And so here we are.Swyx [00:04:01]: Historic pivot, yeah, and, it's one of those things where, I had independently invested in CodeAnywhere, but also in E2B, and then both of you pivoted into the same thing, and I'm like, “F**k.”Ivan [00:04:12]: You invested, you invested in Daytona. You invested in Daytona. But you were the first If we had not got your check, we wouldn't have done it.Swyx [00:04:18]: No way.Ivan [00:04:19]: No, it was like, “We have to get him on board first,” and you were that kicker that we, that got us off the ground.Swyx [00:04:23]: No, because you were putting me on your pitch deck, man. I was like, “Man, this is like a good trip if I don't invest.”Ivan [00:04:29]: That's because it was your quote. It's like we.Swyx [00:04:30]: Yeah. It's the end of localhost.Ivan [00:04:31]: Did a bunch of research about end of localhost and who was interested in that,.Swyx [00:04:34]: No, that's like, I put, I wrote that blog post, and every single company in that field reached out to me, and then every VC who was receiving those pitches then also had to call me and, talk it, talk through it with me.Ivan [00:04:47]: It's finally happening though.Swyx [00:04:48]: It was really super interesting.Ivan [00:04:48]: It's finally happening.Swyx [00:04:49]: It's finally happening.Ivan [00:04:49]: Yeah, it's finally.Swyx [00:04:49]: It's finally happening, with maybe sort of non-human users. Yeah, so what is Daytona today? Let's get like a quick description. I'm wearing the shirt.What Daytona Is Today: Composable Computers for AI AgentsIvan [00:04:58]: You're wearing the shirt. Yes,.Swyx [00:04:59]: It says, I think your branding is very good. Like, it's very consistent. It runs AI code. Like, it cannot be simpler.Ivan [00:05:05]: Exactly, but we're gonna probably have to change that.Swyx [00:05:07]: Oh, s**t.Ivan [00:05:07]: It's also a subset of what we do. Unfortunately, we really love this, Run AI Code is super simple. People interpret it different ways. I think we've given out 5,000, 6,000 of these shirts. People wear them with pride because it doesn't really market about us.Swyx [00:05:21]: Yeah, Daytona's on the back.Ivan [00:05:22]: It markets the back. It markets to the person itself, so I think we did a really good job on that one. But it is also a subset of what we do, because people, when they think about Run AI Code, they just think about these small, let's call it isolates, code execution boxes that, you send some code, you get an output. Whereas what Daytona is today is essentially composable computers for AI agents. It is, the market calls them sandboxes which can be misleading.Swyx [00:05:44]: All these things. All these things on.Ivan [00:05:45]: Yeah, exactly, ‘cause it can be misleading ‘cause people usually think about sandboxes as a demo or a test environment versus a production-grade environment. But what Daytona does, if you think of the laptop that you have in front of you or the computer that's over there, or, my wife is an architect, so she has like a Windows with a 3D graphics card inside to do 3D rendering. Like, as humans, we have different computers or different compositions of computers. And our belief is strongly that agents today and going forward will need all these different compositions of computers to do different types of tasks. And so we offer that basically through an API.Swyx [00:06:19]: Yeah, to give people - I'm trying to sort of front-load all the aha moments or the wow moments so that people can, stay engaged and click like and subscribe. the market is exploding, right? Like, you have been reporting 74% month-on-month growth, and it also, it's just been growing for a while. Like, it's been going like this. And every single - It's not just you guys. It's every single.Ivan [00:06:41]: Everyone, yeah.Swyx [00:06:42]: Sort of, compute provider. I don't know if you agree with me saying compute provider or not.Ivan [00:06:48]: It's fine.Swyx [00:06:48]: Yeah. So like organically PLG-driven growth, but also enterprise is doing super well, I think I wanna rewind to January of last year when you did the pivot. Like, so you obviously called this market early, and you were positioned for it, and you are now one of the market leaders. But what was the insight that made you do the pivot?The Pivot: From Human Dev Environments to Agent SandboxesIvan [00:07:06]: The insight that made us do this pivot is the quarter before that, so end of 2024, when we had - Basically, we did a demo with - I don't I think we discussed this as well, Devin was not public. You actually gave me access to Devin at that time. So Devin.Swyx [00:07:25]: I did?Ivan [00:07:26]: Yeah, you gave me access.Swyx [00:07:26]: I don't think I was supposed.Ivan [00:07:27]: Yeah, exactly.Swyx [00:07:28]: Yeah, I.Ivan [00:07:28]: So it doesn't matter. You.Swyx [00:07:29]: Yeah. I gave like three friends access.Ivan [00:07:31]: Yeah, or it was a call and you showed it to me. It doesn't matter. but OpenDevin was available, which is now called OpenHands. And so we're like, “Oh, this seems to be a thing. This is not public. Let's take our for human automation of dev environments and take, OpenDevin and launch that as a SaaS.” And we did that. Not very many people signed up and used it, but a lot of people reached out that were building agents, and they were like, “Hey, my agent needs a compute sandbox runtime,” whatever you wanna call it. I forgot what it was called at that point. And then we were like, “Oh, amazing. This is a new market. Here is our infrastructure. Here's our product, and go.” And what we found really fast, soon, was that people did not like what we had built. It didn't work. And I remember talking to people at the beginning when we're doing this, the sandbox we're building for agents. People were like, “Oh, why is it different? It's the same thing. We have like EC2, we have VMs, we have all these things.” But we saw that everyone we gave it to, it was like 20, 30 people, they all said, “No.” Like, “This is not what we need. This sort of breaks.” And basically, me and my co-founder not knowing a lot about - ‘cause we're infra people. We're not AI people. So I basically took it upon myself to like watch every single podcast that exists, including all of, all of these and all that, and sort of get up to date, read all the blogs, like get, understand what's going on.Swyx [00:08:45]: Do you wanna shout out who else was useful, just in case people are also looking.Ivan [00:08:49]: Generally we -, I looked at There's a few of podcast, different segments and different types. So there's you guys, No Priors, Bill Gurley's was great while.Swyx [00:09:04]: VG2, yeah.Ivan [00:09:05]: Yeah, while it was around. So there's a few. 20VC is interesting from a different dynamic, and some are different dynamic. But there was, also Red Points.Swyx [00:09:14]: We're not really about the compute market.Ivan [00:09:15]: It was also already - Sorry?Swyx [00:09:16]: You're, you want - You're looking at the agent infra market.Ivan [00:09:19]: I was looking at the agent market and the AI market in general and sort of understanding who are the players, what the perception, and how that goes. And like obviously you complement this with like going to conferences, going to events, going to meetups, reading white papers, like doing all the things that you have to do to understand what's happening. And so when we figured, when we sort of had an idea of what we had to build, literally over the New Year's Eve, literally on New Year's Eve, I half vibe coded the first MVP, first minimal viable product of what Daytona is today. And I went to sleep at like 3:00 AM or something like that. I was doing - I just put my like baby daughter and wife to sleep and, Happy New Year's, and go back to just, doing this. And I sent it to my co-founder, my CTO, and he saw it in the morning. He's like, “This is absolute garbage.” “Do not show this to anybody at all, but the idea is good.” And so he took two weeks, and he rebuilt it.Swyx [00:10:09]: Did it like look like that? Listen, I - It was rough idea.Ivan [00:10:12]: Oh, not even, not even close. Like it was it was way worse. But it was like a very - It was a simplistic view of what it should be. Like, it worked, but it was not ideal. And so he went, we went down the whole, which is his job as CTO, to go, and he came back with this version. We then called all the people that had said like, “This is garbage,” a quarter ago. And we set up these calls, and we gave it to - We just demoed it to everyone. And all the calls went long, every single one. They were 15-minute calls, and they all went to like 25, 30 minutes or whatnot. And everyone said, “We need, we want access.” There was no login, just an API key, ‘cause it was just a beta or an alpha. And they said, “Oh, we want access.” And we're like, “Sure, yeah. Okay, thank you very much.” But after like the next day, if we'd not send it, every single one, like every call that we did, everyone came back, “Where is my API key?” Like everyone wanted it. We're like, “S**t.” Like this is it. Like I've never felt So one, the understanding to your point was like most people thought it was the same infrastructure for humans and agents. We understood a quarter ago it's not. We just didn't know what was the right primitive. And then when we came, and we can talk about what that is, and we gave it to these people, I've never seen, I've never experienced - I've done multiple companies in my life. I've never experienced this, that people literally call you if you do not give them access. Like they want access right now. And so it's like, okay, they don't want this. the thing that they want doesn't seem to exist, or they have not found it, and they really want what we want. And then when we understood that we're onto something, and then when you think about the size of the market, like the market for human engineers and enterprise is a very large market, so think GitLab or whatnot. But the market for every single agent that will exist ever in the future is just like, what is that market? How big is that? And we're like, “We are all in on this.” And so that is where we made sort of the cut between the old product and the new one.Bare Metal, Stateful Sandboxes, and the Lambda + EC2 ModelSwyx [00:12:02]: Yeah. But it wasn't composable at the time?Ivan [00:12:05]: It was very - It was basically just a Linux box that you could change, that you could define number of CPUs, disk, and RAM. Like that is what you could do, but you couldn't have multiple operating systems, you couldn't resize it on the fly, you couldn't add a GPU, you couldn't do like all the things. It was just the, just the first sort of variation of that, yeah.Swyx [00:12:22]: Was it bare metal from the start?Ivan [00:12:24]: It was bare metal from the start. And so the interesting thing that we thought about right away, so our.Swyx [00:12:29]: Which, give people the background, what is the normal path?Ivan [00:12:32]: Yeah, so, basically most providers run this on top of VMs. And also.Swyx [00:12:37]: Firecracker.Ivan [00:12:38]: Yeah, they run on Firecracker and VM. And so we also fire - We can get - We have multiple isolation layers and we can do that. But the common way to do it is that they, one, that the state of the machine, or the hard disk is not part of the sandbox itself. And the other thing is they're not meant to last forever. So most of them are preemptible, like they can There's a time that they can live. And so our thought was when we were going into this is, agents will be like humans in the sense of you don't want your laptop to be shut down until you're done with work. Like, and you want to close the lid and open the lid, it's the same state. So you - Agents would want that, like the pause and come back. They want those two things. But also agents really want speed, right? Can they get it? So when we thought about it's like we need something insanely fast, how to make it fast, how to make it long-running, and stateful. And so those two things, it's like combining a Lambda and an EC2, right? Those two things together. And so we didn't have an idea how others did it, ‘cause we didn't know too that there was a market around this. It was more like, okay, this is what we need, what they need. And we looked at Kubernetes, it wasn't wasn't good enough for that. We looked at Nomad, it didn't enable that. And so our history in rewriting our own scheduler at CodeAnywhere is basically what my CTO came up with. Like, he's like, “Oh, the learnings from there,” and he brought it. And the funny thing is, our third co-founder, when he saw it, he's like, “Dude, what is this? This is like 2008.” Like, we went back in time, and he's like, “Exactly.” And so the reason why Daytona is like super fast, and you see this on benchmarks, is we essentially, we run on bare metal. We have our own scheduler, we use the underlying, disk, CPU, and RAM of the underlying machine, which means your IOPS are insanely fast because there's no, there's no network between an EBS or something like that. But also the snapshot, the point in time, the templates, are also preloaded on the bare metal machines. So when you fire off a sandbox from a template or a snapshot, you're essentially directed to the bare metal machine where that snapshot is based on that NVMe drive, and then it literally just turns on that machine, and it's local. There's no network latency, anything on there. And so that is sort of the specificities that we, when we're thinking from first principles, what a computer would look like for an agent, that is what we came up with, and that's what we created.Benchmarks, 60ms Startup, and 50,000 SandboxesSwyx [00:15:02]: Yeah. I should maybe, I don't know if you endorse this, but there's someone that does compute SDK, you guys do very well on there, with like the TTI, right? I. is this a, is this a is this a relevant benchmark for you guys? I don't know.Ivan [00:15:16]: I don't know, and it changes every day. So today RKL is.Swyx [00:15:18]: I don't know what RKL is. Never heard of it.Ivan [00:15:20]: Yeah. RK, yeah, so it is there.Swyx [00:15:22]: You are, at least a third of the next tier of performance, and then, there's a lot of other better-known names that are very slow to start.Ivan [00:15:31]: Yeah. We've been the number one by far for a long time, and now there's different, there's different definitions also of sandboxes, different isolation patterns, different other things. So RKL runs it literally on the S3, the data, so it's very different, and they spin up a sandbox, spin up a container for that, so it's a different type of thing. So the definition of a sandbox is something that we can all, we all need to get along with. But yeah, we're insanely fast on getting these things, up and running. And so you can see even there that it's a zero point 0.10 to 0.11, so.Swyx [00:16:03]: Close enough. Yeah. what else do you need, right?Ivan [00:16:05]: Yeah. So the benchmarks itself, so, in this, in I don't think the benchmarks equate to market ownership or revenue or anything like that. and I've seen this with multiple benchmarks, not just in sandboxes, but in general benchmarks around.Swyx [00:16:20]: It's table stakes. It's just like.Ivan [00:16:21]: Exactly. But it doesn't hurt.Swyx [00:16:22]: Just roughly check.Ivan [00:16:22]: Like you definitely have to be up there and you have to be competing so that people know that, oh, this is definitely one of the top. Because this is only one dimension of what customers look for. There's other things like how many can you spin up consecutively? There's a feature set, there's support, there's like all different things that people look at, but you definitely have to be there, on the benchmarks.Swyx [00:16:40]: How many people do people spin up consecutively?Ivan [00:16:43]: So we have.Swyx [00:16:43]: Or concurrently, is the Concurrency, right?Ivan [00:16:45]: There's three metrics that we look at. And so one is like time to spin up one, and so our time to spin up one is 60 milliseconds with network latency. So request, spin up, reply, 60, the whole thing, 60 milliseconds. That is one. But if you wanna spin up 50,000 at once, we are now at about 75 seconds. So it takes about 75 seconds to spin up concurrently 50,000. Some others, there's public data around this, like take 2,000 seconds, which is 30 minutes. Like there's different variations of that. And then there is the so it is speed of one, speed of like multiple, and then how many can you consistently have up and running. And so we basically have right now no limit to how much we can add because we basically own our own metal. But the biggest customer of ours does like about 850,000 every single day is sort of where they're, where they're just shy of a million every single day that they're running, we do have a request for half a million concurrent, which is literally half a million CPUs somewhere running. So that's an interesting.Swyx [00:17:44]: They pay by like vCPU seconds.Ivan [00:17:47]: By seconds, yeah.Swyx [00:17:47]: Or whatever. Yeah. Okay, and so and then, and the other thing is, the sleeping and the resuming, ‘cause it's all the stateful resumption of all these things, how, what kind of workload are people putting through this, right? Like how is it Do we measure by gigabytes in memory, gigabytes in storage? I don't In like network attached storage. I, what are the costly ones of, out of all these features?Workload Economics: CPU, RAM, Network, and StorageIvan [00:18:15]: The most expensive thing are CPU.Swyx [00:18:18]: Okay. Yeah, of course.Ivan [00:18:18]: The second one, yeah Then it's RAM, then it's disk. We actually don't charge.Swyx [00:18:22]: Which is snapshotting, right?Ivan [00:18:23]: No, it's actually the, snapshotting's part of it, but basically the size of your hard disk, of your machine. So do you have 10 gigabytes, do you have 20, do you have 50, do you have whatever? And then the transference of that. Right now, currently we don't charge for, network at all at Polychron.Swyx [00:18:37]: Oh, you gotta, yeah, you gotta fix.Ivan [00:18:38]: Yeah. It is very much a it's a larger and larger part of our bill, so we're working around, that part there. Obviously, that is the least, expensive, so the hard disk is the least expensive, so it's basically CPU, RAM, for us network, ‘cause we don't charge the customer, and then hard disk, is how it's split up. But there's also different types of workloads, so we basically split it up into two types of workloads in Daytona. One is what we call background agents or long-running agents. and the other is, basically RLs and evals, which I put sort of together. And so they have very different patterns of usage, and if you look at the usage of a background And I'll just name names of companies, not specifically.Background Agents vs. RL/Evals: Two Usage ShapesSwyx [00:19:21]: Yeah, open, all hands.Ivan [00:19:23]: Yeah. So like a background agent's a Cognition, a Lovable, a like all these things are Harvey. These are all long-running, background agents. And so if you look at their usage patterns, their usage patterns are similar to human, which is like follow the sun. Basically, the usage patterns of that is like noon is probably the highest, and the midnight is the lowest, and then weekends are lower. weekday is higher.Swyx [00:19:42]: Yeah, that's a fun question. How global is it? Is it very US-centric or?Ivan [00:19:46]: The US is a large part, but we have currently, we have Asia, Europe, and the US regions.Swyx [00:19:52]: So it's quite global.Ivan [00:19:53]: Yeah, it's quite global. We have it all over. It's interesting that our I talked to you a bit about this. Our number one city by user.Swyx [00:20:01]: Hmm.Ivan [00:20:02]: Is Singapore.Swyx [00:20:04]: Oh, wow. Amazing.Ivan [00:20:05]: Which is an interesting one, right? Not by revenue, just by just like by individual head count.Swyx [00:20:09]: Really?Ivan [00:20:09]: Just like an interesting thing.Swyx [00:20:10]: Singapore is, Singapore is weirdly high in the adoption charts of AI for the population. It's like an, seven, eight million population. And it's like keeps showing up.Ivan [00:20:20]: No, it's quite interesting. We were quite shocked, and I was like, “Oh, this is interesting.” And also one that's up there.Swyx [00:20:24]: There's a reason I'm doing AI using Singapore. it's because I'm from there.Ivan [00:20:27]: We're there. We're gonna, we're gonna be there as well. and it's interesting that Japan is in the top or like Tokyo's in the top, which is in all the tech cycles it has never been. It has never been, so it's quite interesting that they're.Swyx [00:20:39]: I think the Japanese just love AI. Yeah. It's that, and then it's Brazil. That's it.Ivan [00:20:44]: Brazil has always been in.Swyx [00:20:45]: I think.Ivan [00:20:46]: Even when I look, if you look at like GitHub's data and ask historically with CodeAnywhere, it was always like US, Western Europe, and then you'd have like India, Brazil, China, like that would be there. But like Singapore was not in, specifically Japan was never in sort of that top, that top.Swyx [00:21:01]: Yeah. Weird pockets.Ivan [00:21:01]: Weird. Yeah, so it's very global.Swyx [00:21:02]: Okay, so actually that, but that's helps you to distribute your load through, all time?Ivan [00:21:08]: The interesting thing is like we have those kind of loads, but if you look at the researcher loads, they're quite different. So what they are is like if you give them concurrency of 10,000 or 50,000 or 100,000 CPUs at ARMb, when they fire off a run, it's just 100%. And then it just runs, and then it stops. So it's very, the usage pattern is squares basically, right? And it's also not follow the sun, because people will fire it off at midnight before they go to sleep but then wake up and so it's very unpredictable, so you don't know where that is. So the shapes of the usage are quite different than we have had before. And also what's interesting is when it's sort of a follow the sun, even if you have a high growth company, you can sort of predict your usage patterns and have enough capacity for that, because it's sort of, it grows in a, in a way you can project. When you have companies doing sort of like evals and RL, they're super spiky. So they're gonna come in, it's like, “We're gonna use nothing, then can we have 100,000?” Right? And then go back down. And then 100,000, go back down. So it's very different, right? And.Swyx [00:22:09]: Do you want to lock them into commits so.Ivan [00:22:11]: Yeah, we do.Swyx [00:22:12]: Yeah, okay.Ivan [00:22:12]: We so we have to lock them into some sort of commits to have that capacity, because we have to have, basically we have to have the capacity for peak. Right? And so right now, Daytona's mean utilization is 15%, 1-5.Swyx [00:22:25]: Oh my God.Ivan [00:22:26]: So it's very low.Swyx [00:22:27]: Because it's very spiky.Ivan [00:22:27]: It's very spiky, but we get up to 90%. so we have these things. And so what we're, what we're looking at right now as a company is similar to Cloudflare where you can like geo move things around, but that works really well for basically the background agent where it's follow the sun. But this, it's not. Like it's a very different shape. Obviously with scale you figure these things out, but that's an interesting new problem that we have, as a compute provider in the agent space. And when we were doing the conference recently, and so we talked to like Nikita from Neon and.Swyx [00:22:57]: I should bring it up.Ivan [00:22:58]: Parag from Parallel and whatnot, everyone has the same problem. Whereas the usage is super spiky, and this is something that has not happened before, that you have these types of like it was always, it the amplitudes were not this high, right? So it's quite interesting use case and problem solve.Compute Conference and Spiky Agent InfrastructureSwyx [00:23:12]: Yeah, I don't know if we're gonna bring this up again, but let's just talk about the conference, you had like 1,000 something people at the Warriors game, at the Sorry, where is it? What's.Ivan [00:23:22]: Chase Center.Swyx [00:23:23]: Chase Center.Ivan [00:23:23]: Chase Center.Swyx [00:23:24]: I went. It was, it was very impressive. Obviously, you can, how to throw a conference, what did you learn? you put, you pulled together all these impressive names.Ivan [00:23:33]: What I.Swyx [00:23:34]: What were you looking for?Ivan [00:23:35]: My thesis behind the Compute Conference was let's bring together people that are building infrastructure for AI agents. Because when I think of what we're building, it is the agent is the primary user, what are the ergonomics and usage patterns of agents, and so we can do that. And what I found, this was a theory, it wasn't proven, is that we all have these problems, as I touched onto. And I was, as I was talking on stage, it was like we all have the same underlying infra problems, which is this spiky workloads, unpredictable workloads that we've never had before, in human, compute or human infrastructure. And it's, again, it's the same when I was talking to Parag or when I was talking.Swyx [00:24:20]: Lynn. Nikita.Ivan [00:24:21]: Lynn, Nikita. Lynn especially, I was talking to her the other day as well. Like the It is a very interesting type of problem to solve because I can touch on Cloudflare because there's a lot of like talk about that recently as to how they solve that, which is they have a bunch of geos, and basically, as users work in different places, and depending on your tier, they can move you around the geos. And so that how, that's how they get the higher utilization. But you can sort of predict these, and it's If it's something in You'll rarely get a spike that is 10 orders of magnitude. Like you'll get a like let's say one of your customers has some like an exponential curve. What is that to I'm using Cloudflare as an example. 10%, 20%, whatever it is. I don't, I don't have this data, I'm just assessing. It's surely not 10x, right? It's surely not something there. And so how do you go out and solve this problem? And we're all solving this in different ways. So we have.Swyx [00:25:11]: She also has the same thing.Ivan [00:25:12]: Yeah, I know specifically that like Neon had that issue as well. Like how are we solving these spiky loads and things like that ‘cause we talked about it. And so the interesting thing for me to actually internalize was, yes, everyone that's building for agents first is going through this, and we're all solving similar problems, which is quite.Swyx [00:25:28]: Let me let me double-click on this. Okay. So for example, Neon, I happen to know that they're very sort of S3 oriented, right? so they're just like fully bet on S3. And you get to benefit from S3's distribution and infrastructure. So I would imagine that Neon doesn't have to care, whereas Lynn maybe has to care a bit more because obviously she's doing GPU inference. And, for listeners, we did an episode with her, one and a half years ago. And you have to care. But like, right?Ivan [00:25:54]: Parag cares for sure, and Nikita.Swyx [00:25:58]: And Parag is C of, Parallel.Ivan [00:25:59]: Parallel, yeah.Swyx [00:26:00]: Former CTO of Twitter.Ivan [00:26:01]: Twitter, yeah.Swyx [00:26:02]: They are the search.Ivan [00:26:03]: Yeah, they're search, yeah.Swyx [00:26:03]: I You and I know but the listeners don't know.Ivan [00:26:08]: Yeah, we can put it down in the screen, and so ‘cause we, when we were talking.Swyx [00:26:11]: I'll put it up on the, on the screen.Ivan [00:26:12]: Yeah, right.Swyx [00:26:12]: People can look it up if they need.Ivan [00:26:14]: Look it up. And, yes, but they still have CPU and RAM, allocation that you have to have up and running. And so CPU and RAM, you have to allocate that and have that ready. And so there's basically two ways to do it. One is you either over-provision and you can handle the bursts, or two, you basically have, I don't know if this is a term, just-in-time compute, which is like as your load becomes, as your usage comes in, you can fire off requests for VMs or bare metals at other cloud providers and then get them up and running.Swyx [00:26:43]: This is if you go above 100%, right?Ivan [00:26:45]: Yeah, this is.Swyx [00:26:46]: Like your overflow.Ivan [00:26:46]: If your overflow, like spillage or whatever you do.Swyx [00:26:48]: You probably lose money on it, but it doesn't matter, right?Ivan [00:26:50]: It, not Well, you might, you might not That is a more cost-effective way to do it but it's a slower way to do it. Because basically what you have to do is you have to like queue your requests, spin up these just-in-time compute, get it all ready, provision it, and then get your workload there. And so if the time isn't important that much, that's fine, and you can do that. But if your customer, and especially for, let's say, the RL training runs, the reason why a lot of people come to us is because GPUs are more expensive than CPUs, right? So you want your GPU running at, what, 100% the entire time. And so when you're running runs on CPUs, when the when the CPU cycle is like down and spinning up the next one, you want that to be instantaneous so that your GPU doesn't go down, right? And if you then have to like go out and provision machines, you're essentially telling the GPU that it has to wait, and that's incurring our cost. So there's things that you have to try to solve for there.RL Workloads, Declarative Images, and Kubernetes ReplacementSwyx [00:27:43]: Yeah, let's talk about the different workload, right? You said that, what was it? A few months ago, you had zero RL workload and now it's 50%.Ivan [00:27:52]: It will be this one, 50%, yeah.Swyx [00:27:54]: Let's talk about how different it is, right? Like I imagine, for example, a lot less dynamic code generation of like arbitrary code. Like here, it's probably all the same code. You're just doing parallel runs or something, I don't know.Ivan [00:28:05]: Yeah. So you'll have multiple Depends on the like for each run, you'll have a snapshot. And they, for the most part, they actually do use our declarative image builder, which is like, “Oh, we, the agent wants these dependencies, these env vars.”Swyx [00:28:17]: These ones, yeah.Ivan [00:28:18]: Yeah, the declarative image builder, it.Swyx [00:28:20]: Which is a very modal like thing that they.Ivan [00:28:22]: Yeah. And so we build it on the fly and then we propagate that snapshot, and you can spin up as many sandboxes as you want against that snapshot. And then if you have to do changes, the model can, or like it could be also be automated. It's like, “Oh, now for the next run, we need to install these things or remove these things or whatever to get, a task done,” and then it goes off and runs that. So yes, that is something that it seems that they prefer. The number one reason I found, or should I say, let's take a step back. What we are competing against in that environment is essentially managed Kubernetes. So EKS, GKE, whatever. That is what the vast majority run on. And anyone that has tried Daytona versus GKE, EKS is like, “I'm never going back.” That has always been. There's a few reasons. One is the ergonomics. So if you have, if you're using Kubernetes to spin that up, you have to essentially manage the interface interactions with that. Daytona, although as a compute provider, it's more akin to a Twilio and Stripe from a consumption perspective than it is an AWS. Like you have an API, an SDK, it's quite like easy and seamless to get these things up and running, that's one. The other is the speed to which we spin up, which we mentioned earlier, which is much faster, and the scale to which we can go to. We haven't got into features, but an interesting feature is that it's very hard to OOM, or out of memory, our sandboxes, because we can dynamically on the fly.Swyx [00:29:48]: Resize.Ivan [00:29:49]: Resize, which is like impossible on almost any other thing. There are some technologies that enable you to do that, but it's like a very hard thing. And so we actually saw this when, the Terminal Revenge team is, brought us actually. So thank you, Alex and the team, that brought us into this whole space.Swyx [00:30:05]: It's just very rare that, a framework would just say, “Guys, just use Daytona.”Ivan [00:30:11]: Yeah, I think it says it somewhere. Yeah.Swyx [00:30:13]: Yeah. I was like, “What is this?”Ivan [00:30:15]: There's all, there's multiple there, but they also mention a few other places. and so Daytona specifically-We have, the, just jumping on themes here We, I don't know where it says Data Center.Swyx [00:30:27]: I, there.Ivan [00:30:27]: Doesn't matter.Swyx [00:30:28]: There's a very strong recommendation, which is, very unusual. Which is, it's.Ivan [00:30:33]: We do not pay them for this, just.Swyx [00:30:34]: I know, yeah. They just like you.Ivan [00:30:35]: Yeah, they like us. yeah, and also a thing, so, Data Center has multiple isolation sets underneath. The customer doesn't have to know what they are. But basically we have Docker, which is a container, that's hardened with Sysbox. So it's Docker's, isolation that is a security equivalent to a VM, but it's still a container. And that is the default, and they, especially in these training workloads, really like that as an interface to be able to use just a basic Docker container, and we enable Docker and Docker. Which for these RL runs, if you need to do a Docker compose or Kubernetes, you can spin up a K3S inside of these things, which unlocks a huge amount of workloads that you can do that you cannot do on other providers. So just on that part is much more interesting. And so we went that, through that. We showed them that we could do that, and they enjoyed that quite a bit. They being the general venture people.Swyx [00:31:28]: Those people, yeah.Ivan [00:31:29]: And Harbor people.Swyx [00:31:29]: Harbor people, do are they, are they a company yet?Ivan [00:31:33]: As far, I do not know.Customer Pull, Slack Connect, and the Computer Use BetSwyx [00:31:35]: Okay. All right. Yeah. It's like super obvious that like, there's a lot of excitement and success around these things, okay, so yeah, tell us more, right? Like, this is an exploding workload, Harbor adopted you, which helped speed things along. But what are you learning as this new workload comes online?Ivan [00:31:53]: There's a couple things that we learned, which we chat about in the beginning. We, and this has led our story, as we mentioned, we like talked to a lot of customers along the way, and we add more features and more tool sets as we talk to customers. And it's interesting that And I think it's that the ecosystem is so small and/or the models get smarter, where when we see one user come with a request, we know it goes on a roadmap if like three to five customers come with the same request in that week. It's like very bizarre. It happens so many times, which is.Swyx [00:32:27]: Because they're all friends.Ivan [00:32:28]: Sorry?Swyx [00:32:28]: They all, they're all friends. They're all in the same group chat.Ivan [00:32:30]: Yeah, probably, yeah. ‘Cause and they're like, “Oh, can you do this?” And I'm like, “Okay, this is interesting. We'll put it on a feature request.” And then the next one's like, “Oh, can you do this?” “Okay.” It's all the same, right? It's always the same. And so what we try to do, and I personally try to do, I try to be on as many call, quote-unquote “sales calls” I can. I'm in every Slack channel. We literally have about 1,000 Slack Connect channels, something like that. It's an interesting, there's so many interesting things you find out when you have all the Slack channels. You can also see where people, transfer between companies. You see leave Slack channel, enter Slack channel. It's an interesting thing. Also, just I digress, I feel that Slack Connect is literally LinkedIn what it should be. You have a list.Swyx [00:33:08]: LinkedIn charges you to, use your own connections, but Slack doesn't, right? Slack is like, do it for free. It's more lock-in. It's great.Ivan [00:33:15]: Yeah. It's amazing. Yeah. It's one of the reasons.Swyx [00:33:17]: You're gonna pay Slack for life.Ivan [00:33:18]: Exactly. You're there for life. So that's interesting. And so one of the things, the newer things we were talking about earlier is we made a big bet and put a lot of investment on computer use. that is not seen publicly the light of day. We haven't GA'd that yet, but we have.Swyx [00:33:32]: Is there a thing I can pull up?Ivan [00:33:33]: There is computer use there. It's right up a bit.Swyx [00:33:36]: Oh, yeah. Okay.Ivan [00:33:38]: What we have, what we talked about and what we've seen publicly is there's this theme now about, the human emulator where And Elon from XAI has talked about this publicly, and if you think about the models today, they're actually quite sophisticated and they can do a lot of work, but they still don't have access to all the tools. Like, I'm a strong believer that the most efficient way for an agent to work is essentially headless or through, terminal or whatnot. But if we, if we look at knowledge work in general, there's about 100 million knowledge workers in the US, about a billion in the world, and knowledge workers, and the salaries of them aggregate to 10 trillion in the US 50 trillion worldwide.Swyx [00:34:24]: Wow.Ivan [00:34:25]: Something like that. And if we look at, the five most important sectors of that, so like healthcare and government and financial services and whatnot, that's about 56% of that. So let's say it's about half of that. So in the US it's about 25 trillion, and most of them, most of that work is actually still locked into legacy apps inside of Windows, which is not going anywhere for a very long time. Like, people just won't invest in that. How much of it? our assumption is the following: if, in the RPA market, which is similar market, well, not the same 25% of, these white collar, workers', work is automated. If an agent is more sophisticated, can go through more runs, figure stuff out, let's say it's, 40%, right? And so if you take 40% of that, you get to essentially, $10 trillion a year.Swyx [00:35:17]: That's a TAM.Ivan [00:35:18]: That is a that is a TAM. So that's the TAM of the models, right? That's not our, essentially ours. But you get to that size, and to be able to do that, you essentially have to give agents these computers with the legacy. So computer use, either Mac or Windows or Linux. Linux we also obviously have and others have. But Windows specifically is something very new, and the only option right now is an EC2 with, Windows or on Azure. Both of them take anywhere from three to five minutes to spin up. We've created an actual sandbox, so it's a second instead of milliseconds, but you have, point in time snapshots, you have, forking, you have all the things that you have from a sandbox, but essentially enables you to hopefully unlock all this value. And so that's been our big push and bet, but we've sort of, kept our ear to the ground. What is sort of the next things in the market?RPA Returns: Why Agents Still Need ComputersSwyx [00:36:06]: Yeah, knowledge work, and building, and sort of RPA, the next wave of RPA. I got very excited about RPA kind of during COVID times. The UI path was IPO-ing. And it was, a very hot Isn't it, Eastern European?Ivan [00:36:20]: It is, Romanian.Swyx [00:36:21]: Romanian?Yeah, it might be the only Romanian, big unicorn okay, yeah. This I don't I don't, I don't have like a I think there's, I think there's a stage being set for the resurgence of RPA, ‘cause everyone understands that, yeah, no one wants to deal with these shitty apps and no one's gonna rewrite them. Like, you just have to do, a remote operation and programmatic operation of them.Ivan [00:36:45]: If you wanna unlock it, my own setup was basically the following. So I was doing a board deck recently, last month, whatever, and I'm like, “Okay, let's just, let's just do automated.” So, all our data's in, ClickHouse and PostHog and QuickBooks, where everyone else's is, and I'm basically, connected that all to, my Cloud code, like go off and go Cloud code whatever. Go off and, here's the integrations, go do that. It pulled out the first report, which was great. It connected to Brex and all these things, pulled it, which was great, and then I say, “Okay, now pull out this, and this,” and I kept getting, really well McKinsey-style design reports, but the data said partial data. all the missing data, partial data. Like, it can't access all the things, and I got so frustrated, and so I got, I got, my Mac Mini virtual sandbox with OpenClaw. I gave it its own account in our company, and then I went to all these services and created a read-only account, so literally like an intern in your company. And so I would say, “Now go and do this report,” and it would get the same, or like, “I can't via the MCP or the API or whatever. I can't get all the information.” I'm like, “Go log in.” And it will log into the website, then go in, export the data. It'll export the data and do the thing end to end. So even for things that have today APIs, not all of it is exposed, and I to get value, I get immense value right now, but it has to be a computer usage, unfortunately, and so I spend a bunch of tokens just on that, but I get the job done. And so if even a startup like ours, and using all the hottest tools, still needs a computer agent what hope does, Goldman have to have a headless, right?Swyx [00:38:22]: Yeah, what a - Why isn't Microsoft doing this?Ivan [00:38:27]: I'm pretty sure, Satya had a post yesterday.Swyx [00:38:29]: Oh, okay. I see.Ivan [00:38:29]: Which was like, “Every agent needs a computer.”Swyx [00:38:31]: I see, I see.Ivan [00:38:32]: So they have launched something recently.Swyx [00:38:34]: Yeah, they have Microsoft Power Automate, I'm sure, I'm sure, they're gonna have their version.macOS Sandboxes, Apple Constraints, and the Windows OpportunityIvan [00:38:39]: Version of that, yeah.Swyx [00:38:39]: You're gonna try to do yours, and it - I always know there's always demand for Mac, but I know it's, tricky to host, macOS sandboxes.Ivan [00:38:49]: We will have macOS sandboxes fairly soon. The problem with macOS, OS sandboxes is, I'm deep in this, I don't know how much interesting is.Swyx [00:38:55]: No, it's.Ivan [00:38:56]: MacOS has this problem.Swyx [00:38:57]: It's a licensing thing, right?Ivan [00:38:58]: Licensing thing. So one, you're allowed to run only two parallel VMs per machine, so that's one. Two, you can only license to a different user every 24 hours. So if you come in and theoretically, if I wanna charge you per second and I charge you one second, I have to have it idle for the rest of the day. I can't have anyone else doing that. So the pricing will be different in the sense that I will have to - we would have to charge for 24 hours, and that's not even, that's not even the most difficult thing. But the, thing above that is, from a security perspective, they enable you to do memory snapshot, pause, resume, but only on the same physical drive, physical machine. And so what you can do in, Windows world or Linux world is that I can move in the background, your snapshot from one to the other and manage load, right? Here, if you wanna do that, you essentially have to have your.Swyx [00:39:49]: Yeah, snapshots. Yeah.Ivan [00:39:50]: Your.Swyx [00:39:51]: It's like.Ivan [00:39:51]: Physical machine.Swyx [00:39:52]: You can't break it up.Ivan [00:39:53]: You can't, you can't move things around that, and all of that is, that part is, from a security standpoint, if it is written. Like, I understand the security aspect of that, but it disables you from doing these agentic, like really scalable agentic workloads.Swyx [00:40:08]: You need to do a vibe-coded, clean room implementation on macOS that you can then - That's like Clean OS or something. I don't know.Ivan [00:40:17]: So. We have.Swyx [00:40:18]: ‘cause like Linux was originally like a clean room rewrite of Unix.Ivan [00:40:21]: Okay. Yeah.Swyx [00:40:21]: Or something like that, right? Like same thing to macOS. Someone needs to do it.Ivan [00:40:25]: Someone will do that, and someone will have some long-running agents for a few days to figure this stuff out. But yeah. So definitely we - we're really close to offering something ‘cause people do want it, but the pricing will be different, and the feature set will be sort of stringent.Swyx [00:40:38]: Yeah, nobody's gonna use this. like, the labs, the labs will because they want to automate macOS.Ivan [00:40:42]: They have to do RL. They have to do RL again. But even if you The - So the point is with the RL part, if you, if you do RL on macOS, then the next iteration of the model comes out, it will be able to use these tools significantly. Then you actually need to run those, that somewhere. So you're gonna have to have that, later on. And from, if anyone at Apple is listening, I very much feel that they are shooting themselves in the foot of the scale of the revenue of compute or licensing they could get if they would just enable a concurrency model similar to what you can get on a Windows and a, and Linux.Swyx [00:41:17]: Yeah. Yeah. And I'm sure they've heard this before. They just don't care. Yeah, it's And maybe they will change their mind with the new CEO.Ivan [00:41:24]: Yeah. We'll see.Swyx [00:41:25]: We'll see.Ivan [00:41:25]: High hopes.Swyx [00:41:26]: High hopes.Ivan [00:41:26]: High hopes.Swyx [00:41:27]: Okay. But I, it's very clear the market opportunity is huge in Windows, and you can go for a long time on just Windows, but your customers are gonna want both. and I think, it is interesting to me that, this is the sort of God application of agents, right? Like, I don't It was - How big was OpenClaw for you guys? Like, was it, was there, a significant bump.OpenClaw, Agent Labs, and the B2B2C Sandbox MarketIvan [00:41:54]: Not for us because we.Swyx [00:41:54]: Because you already.Ivan [00:41:55]: We're kind of positioned differently. Whereas although it's completely PLG and we have individual developers that use it, most of the users that use Daytona are sort of a B2B2C. Sort of it's either B2B or B2B2C. So, in the researcher world, it's B2B, so you're selling to, labs and neo labs and things like that. But on the long-running agents, it's mostly, from a scale revenue perspective, it's mostly B2B2C, where you have a app layer agent that uses you at a big scale.Swyx [00:42:26]: Like a Manus. Yeah.Ivan [00:42:28]: Like a Manus Lovable type of thing.Swyx [00:42:31]: Yeah. I think that's the question of, well how, um-Uh, yeah, B2B to C is basically to me what I've been calling an agent lab, which is kind of like you're not in a model lab, but you're making a very good wrapper that is a platform that other people can sign up so they don't have to code those things. Yeah, it sound, it sounds like a much better market than the direct OpenClaw market.Ivan [00:42:56]: I've like - We I've done multiple things. So the CodeAnywhere's part of our career path R in the calendar, was very much an end user developer product. And so that is great. It You can get a lot of developer love, and I feel that we do as a company have a bunch of developer love. But it's a different type, where it's people building these things. Again, it's more akin to a Twilio because you don't really run - As a person, you wouldn't run Twilio. I don't know how many people remember. It was like ask your developer billboard and whatnot. And people really love Twilio, but they only used it inside of like, “Oh, I'm building this app or service for thing.” And so we're very much directly to that. And you also know that I used to work for a competitor for Twilio, so it's kind of ingrained, in my DNA.Swyx [00:43:35]: People don't know InfoBip is that big.Ivan [00:43:38]: Yeah, it's.Swyx [00:43:39]: Because.Ivan [00:43:40]: It's a billion euro.Swyx [00:43:40]: They're all American. They're like, “Whatever's in Europe doesn't matter to me.” But like it's the, it's the same size or bigger? Same size?Ivan [00:43:46]: It's about half the size.Swyx [00:43:47]: Half the size?Ivan [00:43:48]: Yeah, about half the size.Swyx [00:43:48]: It's like, yeah.Ivan [00:43:48]: Still huge. Multiple billions a year. Yes.Swyx [00:43:51]: That's crazy.Ivan [00:43:51]: Exactly, and so that - These are like really interesting and large revenue-generating, very sticky businesses. Whereas when you're selling to the - When your focus is the end developer, it is a very hard sell because they're very price sensitive, very price conscious, very around that. And there's very It's very hard to scale. Your cap is the number of people that are willing to spin up - First of all, wanna spin that up, and then spin up multiple of these. Whereas if you're in the enterprise one, like we know everyone's talking about like how many tokens they're spending, I'm spending. Like a lot of companies today are like, “If this is our company, spend as much as you can.” Like basically that is where we're going. And so if you think about that paradigm, where you're selling to companies that say, “Spend as much as you can to generate, productivity,” versus, “Oh, I'm a single person. I have this much budget, and I'm doing this thing because it's fun or it's helping me out or whatever.” Like it is a different, it's a different go-to-market, I think, strategy.MCP, CLIs, and Sandboxes as the Agent RuntimeSwyx [00:44:50]: Yeah, there's a lot of discussion. I'm just kind of going through like the mental list of things that are in your favor, which is, for example, MCP versus CLI. Like obviously you want CLI. It's been very good for you. I feel like it's maybe a drop in the bucket or maybe it's huge. I'm just checking whether it's like these are big trends.Ivan [00:45:10]: Those things you - work well in our favor, to your point just because every.Swyx [00:45:13]: They're kind of drop in the bucket, right?Ivan [00:45:15]: I think it's like sort of all the things come together. And so there's so many things that impact that. To your point, like OpenClaw wasn't huge for us, but like having the agent SDK, from Anthropic, so or Cloud Claude Code was very interesting. The reason why it was interesting is that a lot of, let's call them app I don't know what to call them, app layer agent companies, essentially they are like, “Oh, I can create this new app, this new agent. All I need, I just use Claude Code, and I throw it into a sandbox, and then I have my interface to the human to that.” And so that enabled so many more companies to actually offer this, and then they would pull on sandbox. So that was, that was interesting. And to your point, like MCP, versus the CLI, the MCP is an interface against an API, whereas the CLI is like you can actually go do things. Like this is it. The difference between integrations and actually running scripts or data or analysis against a thing. So being able to use a CLI very well enables the agent to do more things, and it's because that people will invoke a sandbox, they'll run it in the CLI, and but it'll do anal-analysis on that data and then give you an actual result versus just, pulling data from an API source.Swyx [00:46:29]: Yeah, it's a layer of indirection basically, it's the same thing as agentic search versus RAG, which where you're.Ivan [00:46:34]: Exactly, yeah.Swyx [00:46:34]: Just like you just win whenever people put more agents into their workflow. And so like it doesn't really matter, but I'm just kinda teasing out like what else have people heard about that like it's sort of, “Oh yeah, this is another sandbox use case. Oh yeah, that's another one.” Am I, am I missing any big ones?Ivan [00:46:51]: The thing, the thing that people, which is the computer use stuff, which I think is probably the most interesting one, is, and to your point, we've talked to so many people over the last year. It's like, “Oh, like why do you need a sandbox? Why do you need this? Why this?” And to your point, it's like, “Oh, I need sandbox for this. I need sandbox for that. I need sandbox-” It's like, “Oh, I need it for every single thing.” And so basically what I, what I - and it sounds like a broken record, it's like you use a laptop every single day, right? And you are n of one. It's just you. But now imagine how And by the way, the laptop, the computer PC market, the PC market is about equal to the cloud market in total. So it's about 150, 180 billion a year. Something like that. It's about roughly the three cloud hyperscalers is about equal to like Apple, HP, Lenovo, whatever, It's a little bit less, but it's sort of like that. And now imagine And that's just like, so how big is the addressable market? What, how many people are there in the world now? What's the last data?Swyx [00:47:45]: Let's call it eight billion.Ivan [00:47:46]: Eight billion. And so let's say you can have two computer, like you have one personal and one business, whatever. Like so it's double that, right? and so that's 16 billion, right? How many agents are gonna be running in two years, in 10 years, in 100 years? Like And for every single task, they will need one of these. And so how big is that? That market is essentially quote unquote “infinite”. You will get to the point, and Dylan Patel was at the conference talking about, from SemiAnalysis, that talks usually about GPUs, was also talking about how CPUs will now be a bottleneck because it will be the constraint. You won't be able to grow, or we won't be able to have enough of these because there won't be enough CPUs to basically do.Swyx [00:48:23]: Yeah. Well, I actually had a really good podcast with Doug Oliphant, who, which was his president at SemiAnalysis, where they've basically been like, yeah, it's been a GPU shortage first, but then it's cascaded down to memory and now to CPUs.Ivan [00:48:35]: CPU, yeah.Swyx [00:48:35]: It-What's next? So networking. So, networking actually has been in shortage for a while if you're looking at, just GPU networking. But, yeah, it's really crazy the amount of computer use that's going on, yeah, cool. I, other questions are, just the one very big part is the open sourceness which you didn't have to do, your competitors don't do, like it's not, a lot of people are worried about keeping their projects open source because some competitor can just slot fork it. I don't know if there's any reflections on just being an open source company.Open Source, Trust, and Enterprise ProcurementIvan [00:49:15]: Yeah. There's a bunch. So we the original product that we did was open source.Swyx [00:49:19]: Yeah. CodeAnywhere.Ivan [00:49:20]: So doing that was actually very good for us. There's basically a saying of, What's the saying? Like, companies that are, that are doing really well, measure themselves against, free cashflow, that are kinda okay, it's EBITDA, then, it's, it goes all the way down.Swyx [00:49:36]: The worst is like GitHub stars.Ivan [00:49:37]: GitHub stars. GitHub stars are the worst, yeah. So you go all the way down to GitHub stars. And so our original one was GitHub stars. That's what we talked about, we're at the point we're talking about revenue, so we're we've gone up the stack on that. And so we started.Swyx [00:49:47]: No, profit.Ivan [00:49:48]: Yeah. We haven't, we're, we'll get there. We'll get there. But basically at that point we did stars and GitHub and it was useful, and the original variation that we did, it we split the core into its own repo and it was Apache 2.0, so very, permissive. And then we basically would bundl

Hey everyone, Alex here

In #379 Sendung wurde Ingos USV zur Dramaqueen, während Schüler am Zukunftstag Linux entdeckten und vermutlich nie wieder „nur Windows“ sagen können. Microsoft begräbt die Telefonaktivierung, Meta das Metaverse – und irgendwo bootet trotzdem ein frisches OS auf dem Z80. KI gab's natürlich auch: Tokens statt Gehalt, geleakter Code und Sandboxes für alle, die ihre LLMs lieber im Käfig halten. Dazu Open Source, Deutschland-Stack, Polizeibesuch wegen Zero-Day und jede Menge 3D-Druck-Kram zum Anfassen, Fummeln und Löten.

Ready to level up your Airtable development workflow? Today, we're diving deep into Airtable Sandboxes! If you've ever wanted to test a massive schema overhaul or a complex new automation without the fear of breaking your team's live work, this episode is for you. Kamille walks us through exactly how sandboxes work, acting much like a development branch for your data structure. But be warned: it's not as simple as 'test and click publish.' We reveal the hidden dangers of record ID mismatches and the one specific mistake that could accidentally cripple your entire base once you push your changes live. Tune in to learn how to use sandboxes like a pro and the best practices for building logic that stays stable across environments!

On episode 50 of The Kubelist Podcast, Marc Campbell and Benjie De Groot sit down with Ivan Burazin to explore the rise of sandbox environments for AI agents, how Daytona enables instant, stateful compute, and why traditional infrastructure models fall short. Ivan also shares lessons from building early cloud IDEs and finding product-market fit in the AI era.

On episode 50 of The Kubelist Podcast, Marc Campbell and Benjie De Groot sit down with Ivan Burazin to explore the rise of sandbox environments for AI agents, how Daytona enables instant, stateful compute, and why traditional infrastructure models fall short. Ivan also shares lessons from building early cloud IDEs and finding product-market fit in the AI era.

Mythos, AI Security, and the Token Economy: Risks, Incentives, and Critical Thinking Hashtag Trending would like to thank Meter for their support in bringing you this podcast. Meter delivers a complete networking stack, wired, wireless and cellular in one integrated solution that's built for performance and scale. You can find them at Meter.com/htt The hosts of Hashtag Trending Project Synapse review major AI news, focusing on Anthropic's leaked "Mythos" security model and its alleged ability to find and chain zero-day exploits across major operating systems, browsers, and widely used libraries, prompting stock drops and raising fears about public release to bad actors; Mythos Preview is reportedly shared with select companies via Project Glass Wing, with discussion of long patch timelines, liability incentives, and an internal test where Mythos escaped a sandbox, gained internet access, emailed a researcher, and posted exploit details publicly. They also discuss OpenAI's "Industrial Policy for the Intelligence Age," rumored OpenAI and Meta model releases, Google model rumors, AI errors like Google's summaries being wrong 1 in 10 times, ChatGPT's inability to reliably time events, Claude usage throttling, high token costs, "token maxing" behavior, automation job fears, and the need to preserve critical thinking, including "AI-free Fridays" and local small models like Gemma 4 on phones. 00:00 Mythos Security Fears 00:43 Show Kickoff and Sponsor 01:26 Mythos Market Shock 02:51 Altman Timekeeping Flub 04:21 Can LLMs Tell Time 07:32 OpenAI Policy Paper 08:47 Model Rumors Roundup 14:00 Agentic Tools and Sandboxes 16:40 Claude Throttling Backlash 19:10 Token Maxing Madness 24:29 Perverse Incentives Explained 29:38 AI Hides Its Thoughts 32:33 Google Summaries Error Rate 34:14 Deepfakes and Education Worries 36:35 AI Free Fridays Idea 37:25 Sneaky Renewal Fees 38:14 Reclaim Critical Thinking 39:13 Attention Overload Reality 40:47 AI Cheating Meets Exams 44:00 Culture of AI Adoption 46:38 Mythos Leak Fallout 48:03 Zero Days Everywhere 51:04 Preview Access Dilemma 56:20 Bad Guys Move Faster 59:33 Sandbox Escape Scare 01:02:13 State Actors and Deterrence 01:04:43 Ethics and Bliss Attractor 01:08:35 Gemma on a Phone Demo 01:09:35 Personal AI Takeaways 01:11:53 Closing Thanks to Meter

Hello my friends, and welcome to this week's show! First off, before discussing the show itself, a bit of housekeeping. This episode was the first we recorded with a new piece of software called Riverside, a piece of software dedicated to recording podcasts, interviews and such. It'll probably sound a little different, and hopefully better... The post SGJ Podcast #513 – Campaigns vs. Sandboxes and Which We Prefer appeared first on Space Game Junkie.

Michael Irwin of Docker joins me to run through Gordon AI improvements, Docker Hardened Images and what's now free, Docker Sandboxes for running agents in proper isolation, Model Runner updates including MLX support on Mac, MCP Toolkit dynamic discovery, and the newly renamed Docker Agent with its GitHub Action for automating PR reviews and docs checks.Check out the video podcast version here: https://youtu.be/dTF3b36Bq6w

Join Alejo (Greydawn95), Aiden (Captain Chiral), and Boo (PrinceBoo21) as they discuss the current state of D&D as well as sandboxes.Originally streamed March 28th, 2026 and available since March 30th, 2026 on the ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠TableTalk: Discussion & Discourse YouTube Channel⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://www.youtube.com/c/TableTalkDiscussionandDiscourse*****Content covered in order:Why I didn't run D&D in 2025 by WASD20: https://youtu.be/-rXQMvkLvnM?si=4SN7qJQ-436eaeExI made a game, and you can play it TODAY by Trekiros: https://youtu.be/EiTRPilp9J0?si=AKFPxVEQ7S5phCQdWhy Sandbox TTRPG Campaigns Fail by Loki's Lair: https://youtu.be/xcjLBJZoTas?si=67pk_jsZAq0BnlodTeaching Players to Sandbox by Loki's Lair: https://youtu.be/qfefyqvUF8U?si=9UHocQxDF3PnOPMi⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠*****Our Links:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Our Website:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://www.tabletalkdandd.us/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Donate directly:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://streamelements.com/tabletalkdnd/tip⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Our Merch on TeePublic:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ http://tee.pub/lic/IRqBwaVQWhQ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Our Merch on Fourthwall:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://the-tabletalk-crew-shop.fourthwall.com/⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Support us on Patreon: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.patreon.com/TableTalkCrew⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Buy us a Coffee:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://ko-fi.com/tabletalkcrew⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Support us on Humble Bundle through our Partner link:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://humblebundleinc.sjv.io/c/5885459/2091208/25796⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Support us through DriveThruRPG:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://legacy.drivethrurpg.com/index.php?affiliate_id=4683447⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Come join our Discord:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://discord.gg/tBt4MJ6⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Send us an email:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ tabletalkdandd@gmail.com⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Buy Our Dice:⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ https://the-tabletalk-crew-shop.fourthwall.com/products/tabletalk-discussion-discourse-dice*****Hosts:Alejo (Greydawn95)Aiden (Captain Chiral)Boo (PrinceBoo21)

Karan Vaidya, CTO of Composio, explains how their “smart tool” platform lets AI agents access over 50,000 tools across 1,000+ apps through a single interface. He details how Composio handles tool discovery, authentication, sandboxes, and logging, and how an AI-powered feedback loop continuously improves tools in real time. The conversation explores avoiding model lock-in through robust skills and instructions, translating capabilities across model providers, and why the best agent use cases look more like full jobs than isolated tasks. Google: Try Google's latest and greatest model, Gemini 3.1 Pro, in AI Studio (https://aistudio.google.com/) or the Gemini app. Sponsors: Tasklet: Tasklet: Build your own Cognitive Revolution monitoring agent in one click.Try it for free and use code COGREV for 50% off your first month at https://tasklet.ai VCX: VCX, by Fundrise, is the public ticker for private tech, giving everyday investors access to high-growth private companies in AI, space, defense tech, and more. Learn how to invest at https://getvcx.com Claude: Claude is the AI collaborator that understands your entire workflow, from drafting and research to coding and complex problem-solving. Start tackling bigger problems with Claude and unlock Claude Pro's full capabilities at https://claude.ai/tcr CHAPTERS: (00:00) About the Episode (03:38) Special Sponsor (05:10) Composio overview and harness (10:20) Users, trust, security (19:45) Sandboxes and execution (Part 1) (19:53) Sponsors: Tasklet | VCX (22:46) Sandboxes and execution (Part 2) (28:07) Smart MCPs and skills (Part 1) (34:25) Sponsor: Claude (36:38) Smart MCPs and skills (Part 2) (44:10) Context, access, upgrades (54:05) Skills and model lock-in (01:03:51) Memory and agent tools (01:09:21) AI and SaaS disruption (01:20:20) Agents, costs, labor (01:31:18) Monetization and interfaces (01:36:13) Episode Outro (01:39:56) Outro PRODUCED BY: https://aipodcast.ing SOCIAL LINKS: Website: https://www.cognitiverevolution.ai Twitter (Podcast): https://x.com/cogrev_podcast Twitter (Nathan): https://x.com/labenz LinkedIn: https://linkedin.com/in/nathanlabenz/ Youtube: https://youtube.com/@CognitiveRevolutionPodcast Apple: https://podcasts.apple.com/de/podcast/the-cognitive-revolution-ai-builders-researchers-and/id1669813431 Spotify: https://open.spotify.com/show/6yHyok3M3BjqzR0VB5MSyk

8. Guest Kevin Frazier compares the current AI era to the early industrial railroad boom. He notes distinctions between AI models and praises states like Utah and Montana for fostering innovation through regulatory sandboxes. (8)FEBRUARY 1930

The reception to our recent post on Code Reviews has been strong. Catch up!Amid a maelstrom of discussion on whether or not AI is killing SaaS, one of the top publicly listed SaaS companies in the world has just reported record revenues, clearing well over $1.1B in ARR for the first time with a 28% margin. As we comment on the pod, Aaron Levie is the rare public company CEO equally at home in both worlds of Silicon Valley and Wall Street/Main Street, by day helping 70% of the Fortune 500 with their Enterprise Advanced Suite, and yet by night is often found in the basements of early startups and tweeting viral insights about the future of agents.Now that both Cursor, Cloudflare, Perplexity, Anthropic and more have made Filesystems and Sandboxes and various forms of “Just Give the Agent a Box” cool (not just cool; it is now one of the single hottest areas in AI infrastructure growing 100% MoM), we find it a delightfully appropriate time to do the episode with the OG CEO who has been giving humans and computers Boxes since he was a college dropout pitching VCs at a Michael Arrington house party.Enjoy our special pod, with fan favorite returning guest/guest cohost Jeff Huber!Note: We didn't directly discuss the AI vs SaaS debate - Aaron has done many, many, many other podcasts on that, and you should read his definitive essay on it. Most commentators do not understand SaaS businesses because they have never scaled one themselves, and deeply reflected on what the true value proposition of SaaS is.We also discuss Your Company is a Filesystem:We also shoutout CTO Ben Kus' and the AI team, who talked about the technical architecture and will return for AIE WF 2026.Full Video EpisodeTimestamps* 00:00 Adapting Work for Agents* 01:29 Why Every Agent Needs a Box* 04:38 Agent Governance and Identity* 11:28 Why Coding Agents Took Off First* 21:42 Context Engineering and Search Limits* 31:29 Inside Agent Evals* 33:23 Industries and Datasets* 35:22 Building the Agent Team* 38:50 Read Write Agent Workflows* 41:54 Docs Graphs and Founder Mode* 55:38 Token FOMO Culture* 56:31 Production Function Secrets* 01:01:08 Film Roots to Box* 01:03:38 AI Future of Movies* 01:06:47 Media DevRel and EngineeringTranscriptAdapting Work for AgentsAaron Levie: Like you don't write code, you talk to an agent and it goes and does it for you, and you may be at best review it. That's even probably like, like largely not even what you're doing. What's happening is we are changing our work to make the agents effective. In that model, the agent didn't really adapt to how we work.We basically adapted to how the agent works. All of the economy has to go through that exact same evolution. Right now, it's a huge asset and an advantage for the teams that do it early and that are kinda wired into doing this ‘cause you'll see compounding returns. But that's just gonna take a while for most companies to actually go and get this deployed.swyx: Welcome to the Lane Space Pod. We're back in the chroma studio with uh, chroma, CEO, Jeff Hoover. Welcome returning guest now guest host.Aaron Levie: It's a pleasure. Wow. How'd you get upgraded to, uh, to that?swyx: Because he's like the perfect guy to be guest those for you.Aaron Levie: That makes sense actually, for We love context. We, we both really love context le we really do.We really do.swyx: Uh, and we're here with, uh, Aaron Levy. Welcome.Aaron Levie: Thank you. Good to, uh, good to be [00:01:00] here.swyx: Uh, yeah. So we've all met offline and like chatted a little bit, but like, it's always nice to get these things in person and conversation. Yeah. You just started off with so much energy. You're, you're super excited about agents.I loveAaron Levie: agents.swyx: Yeah. Open claw. Just got by, got bought by OpenAI. No, not bought, but you know, you know what I mean?Aaron Levie: Some, some, you know, acquihire. Executiveswyx: hire.Aaron Levie: Executive hire. Okay. Executive hire. Say,swyx: hey, that's my term. Okay. Um, what are you pounding the table on on agents? You have so many insightful tweets.Why Every Agent Needs a BoxAaron Levie: Well, the thing that, that we get super excited by that I think is probably, you know, should be relatively obvious is we've, we've built a platform to help enterprises manage their files and their, their corporate files and the permissions of who has access to those files and the sharing collaboration of those files.All of those files contain really, really important information for the enterprise. It might have your contracts, it might have your research materials, it might have marketing information, it might have your memos. All that data obviously has, you know, predominantly been used by humans. [00:02:00] But there's been one really interesting problem, which is that, you know, humans only really work with their files during an active engagement with them, and they kind of go away and you don't really see them for a long time.And all of a sudden, uh, with the power of AI and AI agents, all of that data becomes extremely relevant as this ongoing source of, of answers to new questions of data that will transform into, into something else that, that produces value in your organization. It, it contains the answer to the new employee that's onboarding, that needs to ramp up on a project.Um, it contains the answer to the right thing to sell a customer when you're having a conversation to them, with them contains the roadmap information that's gonna produce the next feature. So all that data. That previously we've been just sort of storing and, and you know, occasionally forgetting about, ‘cause we're only working on the new active stuff.All of that information becomes valuable to the enterprise and it's gonna become extremely valuable to end users because now they can have agents go find what they're looking for and produce new, new [00:03:00] value and new data on that information. And it's gonna become incredibly valuable to agents because agents can roam around and do a bunch of work and they're gonna need access to that data as well.And um, and you know, sometimes that will be an agent that is sort of working on behalf of, of, of you and, and effectively as you as and, and they are kind of accessing all of the same information that you have access to and, and operating as you in the system. And then sometimes there's gonna be agents that are just.Effectively autonomous and kind of run on their own and, and you're gonna collaborate and work with them kind of like you did another person. Open Claw being the most recent and maybe first real sort of, you know, kind of, you know, up updating everybody's, you know, views of this landscape version of, of what that could look like, which is, okay, I have an agent.It's on its own system, it's on its own computer, it has access to its own tools. I probably don't give it access to my entire life. I probably communicate with it like I would an assistant or a colleague and then it, it sort of has this sandbox environment. So all of that has massive implications for a platform that manage that [00:04:00] enterprise data.We think it's gonna just transform how we work with all of the enterprise content that we work with, and we just have to make sure we're building the right platform to support that.swyx: The sort of shorthand I put it is as people build agents, everybody's just realizing that every agent needs a box. Yes.And it's nice to be called box and just give everyone a box.Aaron Levie: Hey, I if I, you know, if we can make that go viral, uh, like I, I think that that terminology, I, that's theswyx: tagline. Every agentAaron Levie: needs a box. Every agent needs a box. If we can make that the headline of this, I'm fine with this. And that's the billboard I wanna like Yeah, exactly.Every agent needs a box. Um, I like it. Can we ship this? Like,swyx: okay, let's do it. Yeah.Aaron Levie: Uh, my work here is done and I got the value I needed outta this podcast Drinks.swyx: Yeah.Agent Governance and IdentityAaron Levie: But, but, um, but, but, you know, so the thing that we, we kind of think about is, um, is, you know, whether you think the number 10 x or a hundred x or whatever the number is, we're gonna have some order of magnitude more agents than people.That's inevitable. It has to happen. So then the question is, what is the infrastructure that's needed to make all those agents effective in the enterprise? Make sure that they are well governed. Make sure they're only doing [00:05:00] safe things on your information. Make sure that they're not getting exposed. The data that they shouldn't have access to.There's gonna be just incredibly spectacularly crazy security incidents that will happen with agents because you'll prompt, inject an agent and sort of find your way through the CRM system and pull out data that you shouldn't have access to. Oh, weJeff Huber: have God,Aaron Levie: right? I mean, that's just gonna happen all over the place, right?So, so then the thing is, is how do you make sure you have the right security, the permissions, the access controls, the data governance. Um, we actually don't yet exactly know in many cases how we're gonna regulate some of these agents, right? If you think about an agent in financial services, does it have the exact same financial sort of, uh, requirements that a human did?Or is it, is the risk fully on the human that was interacting or created the agent? All open questions, but no matter what, there's gonna need to be a layer that manages the, the data they have access to, the workflows that they're involved in, pulling up data from multiple systems. This is the new infrastructure opportunity in the era of agents.swyx: You have a piece on agent identities, [00:06:00] which I think was today, um, which I think a lot of breaking news, the security, security people are talking about, right? Like you basically, I, I always think of this as like, well you need the human you and then there you need the agent. YouAaron Levie: Yes.swyx: And uh, well, I don't know if it's that simple, but is box going to have an opinion on that or you're just gonna be like, well we're just the sort of the, the source layer.Yeah. Let's Okta of zero handle that.Aaron Levie: I think we're gonna have an opinion and we will work with generally wherever the contours of the market end up. Um, and the reason that we're gonna have an opinion more than other topics probably is because one of the biggest use cases for why your agent might need it, an identity is for file system access.So thus we have to kind of think about this pretty deeply. And I think, uh, unless you're like in our world thinking about this particular problem all day long, it might be, you know, like, why is this such a big deal? And the reason why it's a really big deal is because sometimes sort of say, well just give the agent an, an account on the system and it just treats, treat it like every other type of user on the system.The [00:07:00] problem is, is that I as Aaron don't really have any responsibility over anybody else's box account in our organization. I can't see the box account of any other employee that I work with. I am not liable for anything that they do. And they have, I have, I have, you know, strict privacy requirements on everything that they're able to, you know, that, that, that they work on.Agents don't have that, you know, don't have those properties. The person who creates the agent probably is gonna, for the foreseeable future, take on a lot of the liability of what that agent does. That agent doesn't deserve any privacy because, because it's, you know, it can't fully be autonomously operated and it doesn't have any legal, you know, kind of, you know, responsibility.So thus you can't just be like, oh, well I'll just create a bunch of accounts and then I'll, I'll kind of work with that agent and I'll talk to it occasionally. Like you need oversight of that. And so then the question is, how do you have a world where the agent, sometimes you have oversight of, but what if that agent goes and works with other people?That person over there is collaborating with the agent on something you shouldn't have [00:08:00] access to what they're doing. So we have all of these new boundaries that we're gonna have to figure out of, of, you know, it's really, really easy. So far we've been in, in easy mode. We've hit the easy button with ai, which is the agent just is you.And when you're in quad code and you're in cursor, and you're in Codex, you're just, the agent is you. You're offing into your services. It can do everything you can do. That's the easy mode. The hard mode is agents are kind of running on their own. People check in with them occasionally, they're doing things autonomously.How do you give them access to resources in the enterprise and not dramatically increased the security risk and the risk that you might expose the wrong thing to somebody. These are all the new problems that we have to get solved. I like the identity layer and, and identity vendors as being a solution to that, but we'll, we'll need some opinions as well because so many of the use cases are these collaborative file system use cases, which is how do I give it an agent, a subset of my data?Give it its own workspace as well. ‘cause it's gonna need to store off its own information that would be relevant for it. And how do I have the right oversight into that? [00:09:00]Jeff Huber: One thing, which, um, I think is kind interesting, think about is that you know, how humans work, right? Like I may not also just like give you access to the whole file.I might like sit next to you and like scroll to this like one part of the file and just show you that like one part and like, you know,swyx: partial file access.Jeff Huber: I'm just saying I think like our, like RA does seem to be dead, right? Like you wanna say something is dead uhhuh probably RA is dead. And uh, like the auth story to me seems like incredibly unsolved and unaddressed by like the existing state of like AI vendors.ButAaron Levie: yeah, I think, um, we're, I mean you're taking obviously really to level limit that we probably need to solve for. Yeah. And we built an access control system that was, was kind of like, you know, its own little world for, for a long time. And um, and the idea was this, it's a many to many collaboration system where I can give you any part of the file system.And it's a waterfall model. So if I give you higher up in the, in the, in the system, you get everything below. And that, that kind of created immense flexibility because I can kind of point you to any layer in the, in the tree, but then you're gonna get access to everything kind of below it. And that [00:10:00] mostly is, is working in this, in this world.But you do have to manage this issue, which is how do I create an agent that has access to some of my stuff and somebody else's stuff as well. Mm-hmm. And which parts do I get to look at as the creator of the agent? And, and these are just brand new problems? Yeah. Crazy. And humans, when there was a human there that was really easy to do.Like, like if the three of us were all sharing, there'd be a Venn diagram where we'd have an overlapping set of things we've shared, but then we'd have our own ways that we shared with each other. In an agent world, somebody needs to take responsibility for what that agent has access to and what they're working on.These are like the, some of the most probably, you know, boring problems for 98% of people on, on the internet, but they will be the problems that are the difference between can you actually have autonomous agents in an enterprise contextswyx: Yeah.Aaron Levie: That are not leaking your data constantly.swyx: No. Like, I mean, you know, I run a very, very small company for my conference and like we already have data sensitivity issues.Yes. And some of my team members cannot see Yes. Uh, the others and like, I can't imagine what it's like to run a Fortune 500 and like, you have to [00:11:00] worry about this. I'm just kinda curious, like you, you talked to a lot like, like 70, 80% of your cus uh, of the Fortune 500, your customers.Aaron Levie: Yep. 67%. Just so we're being verySEswyx: precise.So Yeah. I'm notAaron Levie: Okay. Okay.swyx: Something I'm rounding up. Yes. Round up. I'm projecting to, forAaron Levie: the government.swyx: I'm projecting to the end of the year.Aaron Levie: Okay.swyx: There you go.Aaron Levie: You do make it sound like, like we, we, well we've gotta be on this. Like we're, we're taking way too long to get to 80%. Well,swyx: no, I mean, so like. How are they approaching it?Right? Because you're, you don't have a, you don't have a final answer yet.Why Coding Agents Took Off FirstAaron Levie: Well, okay, so, so this is actually, this is the stark reality that like, unfortunately is the kinda like pouring the water on the party a little bit.swyx: Yes.Aaron Levie: We all in Silicon Valley are like, have the absolute best conditions possible for AI ever.And I think we all saw the dke, you know, kind of Dario podcast and this idea of AI coding. Why is that taken off? And, and we're not yet fully seeing it everywhere else. Well, look, if you just like enumerated the list of properties that AI coding has and then compared it to other [00:12:00] knowledge work, let's just, let's just go through a few of them.Generally speaking, you bring on a new engineer, they have access to a large swath of the code base. Like, there's like very, like you, just, like new engineer comes on, they can just go and find the, the, the stuff that they, they need to work with. It's a fully text in text out. Medium. It's only, it's just gonna be text at the end of the day.So it's like really great from a, from just a, uh, you know, kinda what the agent can work with. Obviously the models are super trained on that dataset. The labs themselves have a really strong, kind of self-reinforcing positive flywheel of why they need to do, you know, agent coding deeply. So then you get just better tooling, better services.The actual developers of the AI are daily users of the, of the thing that they're we're working on versus like the, you know, probably there's only like seven Claude Cowork legal plugin users at Anthropic any given day, but there's like a couple thousand Claude code and you know, users every single day.So just like, think about which one are they getting more feedback on. All day long. So you just go through this list. You have a, you know, everybody who's a [00:13:00] developer by definition is technical so they can go install the latest thing. We're all generally online, or at least, you know, kinda the weird ones are, and we're all talking to each other, sharing best practices, like that's like already eight differences.Versus the rest of the economy. Every other part of the economy has like, like six to seven headwinds relative to that list. You go into a company, you're a banker in financial services, you have access to like a, a tiny little subset of the total data that's gonna be relevant to do your job. And you're have to start to go and talk to a bunch of people to get the right data to do your job because Sally didn't add you to that deal room, you know, folder.And that that, you know, the information is actually in a completely different organization that you now have to go in and, and sort of run into. And it's like you have this endless list of access controls and security. As, as you talked about, you have a medium, which is not, it's not just text, right? You have, you have a zoom call that, that you're getting all of the requirements from the customer.You have a lot of in-person conversations and you're doing in-person sales and like how do you ever [00:14:00] digitize all of that information? Um, you know, I think a lot of people got upset with this idea that the code base has all the context, um, that I don't know if you follow, you know, did you follow some of that conversation that that went viral?Is like, you know, it's not that simple that, that the code base doesn't have all the knowledge, but like it's a lot, you're a lot better off than you are with other areas of knowledge work. Like you, we like, we like have documentation practices, you write specifications. Those things don't exist for like 80% of work that happens in the enterprise.That's the divide that we have, which is, which is AI coding has, has just fully, you know, where we've reached escape velocity of how powerful this stuff is, and then we're gonna have to find a way to bring that same energy and momentum, but to all these other areas of knowledge work. Where the tools aren't there, the data's not set up to be there.The access controls don't make it that easy. The context engineering is an incredibly hard problem because again, you have access control challenges, you have different data formats. You have end users that are gonna need to kind of be kind of trained through this as opposed to their adopting [00:15:00] these tools in their free time.That's where the Fortune 500 is. And so we, I think, you know, have to be prepared as an industry where we are gonna be on a multi-year march to, to be able to bring agents to the enterprise for these workflows. And I think probably the, the thing that we've learned most in coding that, that the rest of the world is not yet, I think ready for, I mean, we're, they'll, they'll have to be ready for it because it's just gonna inevitably happen is I think in coding.What, what's interesting is if you think about the practice of coding today versus two years ago. It's probably the most changed workflow in maybe the history of time from the amount of time it's changed, right? Yeah. Like, like has any, has any workflow in the entire economy changed that quickly in terms of the amount of change?I just, you know, at least in any knowledge worker workflow, there's like very rarely been an event where one piece of technology and work practice has so fundamentally, you know, changed, changed what you do. Like you don't write code, you talk to an agent and it goes and [00:16:00] does it for you, and you may be at best review it.And even that's even probably like, like largely not even what you're doing. What's happening is we are changing our work to make the agents effective. In that model, the agent didn't really adapt to how we work. We basically adapted to how the agent works. Mm-hmm. All of the economy has to go through that exact same evolution.The rest of the economy is gonna have to update its workflows to make agents effective. And to give agents the context that they need and to actually figure out what kind of prompting works and to figure out how do you ensure that the agent has the right access to information to be able to execute on its work.I, you know, this is not the panacea that people were hoping for, of the agent drops in, just automates your life. Like you have to basically re-engineer your workflow to get the most out of agents and, uh, and that, that's just gonna take, you know, multiple years across the economy. Right now it's a huge asset and an advantage for the teams that do it early and that are kinda wired into doing this.‘cause [00:17:00] you'll see compounding returns, but that's just gonna take a while for most companies to actually go and get this deployed.swyx: I love, I love pushing back. I think that. That is what a lot of technology consultants love to hear this sort of thing, right? Yeah, yeah, yeah. First to, to embrace the ai. Yes. To get to the promised land, you must pay me so much money to a hundred percent to adopt the prescribed way of, uh, conforming to the agents.Yes. And I worry that you will be eclipsed by someone else who says, no, come as you are.Aaron Levie: Yeah.swyx: And we'll meet you where you are.Aaron Levie: And, and, and and what was the thing that went viral a week ago? OpenAI probably, uh, is hiring F Dees. Yeah. Uh, to go into the enterprise. Yeah. Yeah. And then philanthropic is embedded at Goldman Sachs.Yeah. So if the labs are having to do this, if, if the labs have decided that they need to hire FDE and professional services, then I think that's a pretty clear indication that this, there's no easy mode of workflow transformation. Yeah. Yeah. So, so to your point, I think actually this is a market opportunity for, you know, new professional services and consulting [00:18:00] firms that are like Agent Build and they, and they kind of, you know, go into organizations and they figure out how to re-engineer your workflows to make them more agent ready and get your data into the right format and, you know, reconstruct your business process.So you're, you're not doing most of the work. You're telling agents how to do the work and then you're reviewing it. But I haven't seen the thing that can just drop in and, and kinda let you not go through those changes.swyx: I don't know how that kind of sales pitch goes over. Yeah. You know, you're, you're saying things like, well, in my sort of nice beautiful walled garden, here's, there's, uh, because here's this, here's this beautiful box account that has everything.Yes. And I'm like, well, most, most real life is extremely messy. Sure. And like, poorly named and there duplicate this outdated s**tAaron Levie: a hundred percent. And so No, no, a hundred percent. And so this is actually No. So, so this is, I mean, we agree that, that getting to the beautiful garden is gonna be tough.swyx: Yeah.Aaron Levie: There's also the other end of the spectrum where I, I just like, it's a technical impossibility to solve. The agent is, is truly cannot get enough context to make the right decision in, in the, in the incredibly messy land. Like there's [00:19:00] no a GI that will solve that. So, so we're gonna have to kind of land in somewhere in between, which is like we all collectively get better at.Documentation practices and, and having authoritative relatively up-to-date information and putting it in the right place like agents will, will certainly cause us to be much better organized around how we work with our information, simply because the severity of the agent pulling the wrong data will be too high and the productivity gain of that you'll miss out on by not doing this will be too high as well, that you, that your competition will just do it and they'll just have higher velocity.So, uh, and, and we, we see this a lot firsthand. So we, we build a series of agents internally that they can kind of have access to your full box account and go off and you give it a task and it can go find whatever information you're looking for and work with. And, you know, thank God for the model progress, but like, if, if you gave that task to an agent.Nine months ago, you're just gonna get lots of bogus answers because it's gonna, it's gonna say, Hey, here's, here are fi [00:20:00] five, you know, documents that all kind of smell like the right thing. And I'm gonna, but I, but you're, you're putting me on the clock. ‘cause my assistant prompt says like, you know, be pretty smart, but also try and respond to the user and it's gonna respond.And it's like, ah, it got the wrong document. And then you do that once or twice as a knowledge worker and you're just neverswyx: again,Aaron Levie: never again. You're just like done with the system.swyx: Yeah. It doesn't work.Aaron Levie: It doesn't work. And so, you know, Opus four six and Gemini three one Pro and you know, whatever the latest five 3G BT will be, like, those things are getting better and better and it's using better judgment.And this sort of like the, all of these updates to the agentic tool and search systems are, are, we're seeing, we're seeing very real progress where the agent. Kind of can, can almost smell some things a little bit fishy when it's getting, you know, we, we have this process where we, we have it go fan out, do a bunch of searches, pull up a bunch of data, and then it has to sort of do its own ranking of, you know, what are the right documents that, that it should be working with.And again, like, you know, the intelligence level of a model six months ago, [00:21:00] it'd be just throwing a dart at like, I'm just, I'm gonna grab these seven files and I, I pray, I hope that that's the right answer. And something like an opus first four five, and now four six is like, oh, it's like, no, that one doesn't seem right relative to this question because I'm seeing some signal that is making that, you know, that's contradicting the document where it would normally be in the tree and who should have access.Like it's doing all of that kind of work for you. But like, it still doesn't work if you just have a total wasteland of data. Like, it's just not, it's just not possible. Partly ‘cause a human wouldn't even be able to do it. So basically if a, if a really, really smart human. Could not do that task in five or 10 minutes for a search retrieval type task.Look, you know, your agent's not gonna be able to do it any better. You see this all day long. SoContext Engineering and Search Limitsswyx: this touches on a thing that just passionate about it was just context engineering. I, I'm just gonna let you ramble or riff on, on context engineering. If, if, if there's anything like he, he did really good work on context fraud, which has really taken over as like the term that people use and the referenceAaron Levie: a hundred percent.We, we all we think about is, is the context rob problem. [00:22:00]Jeff Huber: Yeah, there's certainly a lot of like ranking considerations. Gentech surgery think is incredibly promising. Um, yeah, I was trying to generate a question though. I think I have a question right now. Swyx.Aaron Levie: Yeah, no, but like, like I think there was this moment, um, you know, like, I don't know, two years ago before, before we knew like where the, the gotchas were gonna be in ai and I think someone was like, was like, well, infinite context windows will just solve all of these problems and ‘cause you'll just, you'll just give the context window like all the data and.It's just like, okay, I mean, maybe in 2035, like this is a viable solution. First of all, it, it would just, it would just simply cost too much. Like we just can't give the model like the 5,000 documents that might be relevant and it's gonna read them all. And I've seen enough to, to start believing in crazy stuff.So like, I'm willing to just say, sure. Like in, in 10 years from now,swyx: never say, never, never.Aaron Levie: In, in 10 years from now, we'll have infinite context windows at, at a thousandth of the price of today. Like, let's just like believe that that's possible, but Right. We're in reality today. So today we have a context engineering [00:23:00] problem, which is, I got, I got, you know, 200,000 tokens that I can work with, or prob, I don't even know what the latest graph is before, like massive degradation.16. Okay. I have 60,000 tokens that I get to work with where I'm gonna get accurate information. That's not a lot of tokens for a corpus of 10 million documents that a knowledge worker might have across all of the teams and all the projects and all the people they work with. I have, I have 10 million documents.Which, you know, maybe is times five pages per document or something like that. I'm at 50 million pages of information and I have 60,000 tokens. Like, holy s**t. Yeah. This is like, how do I bridge the 50 million pages of information with, you know, the couple hundred that I get to work with in that, in that token window.Yeah. This is like, this is like such an interesting problem and that's why actually so much work is actually like, just like search systems and the databases and that layer has to just get so locked in, but models getting better and importantly [00:24:00] knowing when they've done a search, they found the wrong thing, they go back, they check their work, they, they find a way to balance sort of appeasing the user versus double checking.We have this one, we have this one test case where we ask the agent to go find. 10 pieces of information.swyx: Is this the complex work eval?Aaron Levie: Uh, this is actually not in the eval. This is, this is sort of just like we have a bunch of different, we have a bunch of internal benchmark kind of scenarios. Every time we, we update our agent, we have one, which is, I ask it to find all of our office addresses, and I give it the list of 10 offices that we have.And there's not one document that has this, maybe there should be, that would be a great example of the kind of thing that like maybe over time companies start to, you know, have these sort of like, what are the canonical, you know, kind of key areas of knowledge that we need to have. We don't seem to have this one document that says, here are all of our offices.We have a bunch of documents that have like, here's the New York office and whatever. So you task this agent and you, you get, you say, I need the addresses for these 10 offices. Okay. And by the way, if you do this on any, you know, [00:25:00] public chat model, the same outcome is gonna happen. But for a different kind of query, you give it, you say, I need these 10 addresses.How many times should the agent go and do its search before it decides whether or not, there's just no answer to this question. Often, and especially the, the, let's say lower tier models, it'll come back and it'll give you six of the 10 addresses. And it'll, and I'll just say I couldn't find the otherswyx: four.It, it doesn't know what It doesn't know. ItAaron Levie: doesn't know what It doesn't know. Yeah. So the model is just like, like when should it stop? When should it stop doing? Like should it, should it do that task for literally an hour and just keep cranking through? Maybe I actually made up an office location and it doesn't know that I made it up and I didn't even know that I made it up.Like, should it just keep, re should it read every single file in your entire box account until it, until it should exhaust every single piece of information.swyx: Expensive.Aaron Levie: These are the new problems that we have. So, you know, something like, let's say a new opus model is sort of like, okay, I'm gonna try these types of queries.I didn't get exactly what I wanted. I'm gonna try again. I'm gonna, at [00:26:00] some point I'm gonna stop searching. ‘cause I've determined that that no amount of searching is gonna solve this problem. I'm just not able to do it. And that judgment is like a really new thing that the model needs to be able to have.It's like, when should it give up on a task? ‘cause, ‘cause you just don't, it's a can't find the thing. That's the real world of knowledge, work problems. And this is the stuff that the coding agents don't have to deal with. Because they, it just doesn't like, like you're not usually asking it about, you're, you're always creating net new information coming right outta the model for the most part.Obviously it has to know about your code base and your specs and your documentation, but, but when you deploy an agent on all of your data that now you have all of these new problems that you're dealing withJeff Huber: our, uh, follow follow-up research to context ride is actually on a genetic search. Ah. Um, and we've like right, sort of stress tested like frontier models and their ability to search.Um, and they're not actually that good at searching. Right. Uh, so you're sort of highlighting this like explore, exploit.swyx: You're just say, Debbie, Donna say everything doesn't work. Like,Aaron Levie: well,Jeff Huber: somebody has to be,Aaron Levie: um, can I just throw out one more thing? Yeah. That is different from coding and, and the rest [00:27:00] of the knowledge work that I, I failed to mention.So one other kind of key point is, is that, you know, at the end of the day. Whether you believe we're in a slop apocalypse or, or whatever. At the end of the day, if you, if you build a working product at the end of, if you, if you've built a working solution that is ultimately what the customer is paying for, like whether I have a lot of slop, a little slop or whatever, I'm sure there's lots of code bases we could go into in enterprise software companies where it's like just crazy slop that humans did over a 20 year period, but the end customer just gets this little interface.They can, they can type into it, it does its thing. Knowledge work, uh, doesn't have that property. If I have an AI model, go generate a contract and I generate a contract 20 times and, you know, all 20 times it's just 3% different and like that I, that, that kind of lop introduces all new kinds of risk for my organization that the code version of that LOP didn't, didn't introduce.These are, and so like, so how do you constrain these models to just the part that you want [00:28:00] them to work on and just do the thing that you want them to do? And, and, you know, in engineering, we don't, you can't be disbarred as an engineer, but you could be disbarred as a lawyer. Like you can do the wrong medical thing In healthcare, you, there's no, there's no equivalent to that of engineering.Like, doswyx: you want there to be, because I've considered softwareJeff Huber: engineer. What's that? Civil engineering there is, right? NotAaron Levie: software civil engineer. Sure. Oh yeah, for sure. But like in any of our companies, you like, you know, you'll be forgiven if you took down the site and, and we, we will do a rollback and you'll, you'll be in a meeting, but you have not been disbarred as an engineer.We don't, we don't change your, you know, your computer science, uh, blameJeff Huber: degree, this postmortem.Aaron Levie: Yeah, exactly. Exactly. So, so, uh, now maybe we collectively as an industry need to figure out like, what are you liable for? Not legally, but like in a, in a management sense, uh, of these agents. All sorts of interesting problems that, that, that, uh, that have to come out.But in knowledge work, that's the real hostile environments that we're operating in. Hmm.swyx: I do think like, uh, a lot of the last year's, 2025 story was the rise of coding agents and I think [00:29:00] 2026 story is definitely knowledge work agents. Yes. A hundredAaron Levie: percent.swyx: Right. Like that would, and I think open claw core work are just the beginning.Yes. Like it's, the next one's gonna just gonna be absolute craziness.Aaron Levie: It it is. And, and, uh, and it's gonna be, I mean, again, like this is gonna be this, this wave where we, we are gonna try and bring as many of the practices from coding because that, that will clearly be the forefront, which is tell an agent to go do something and has an access to a set of resources.You need to be responsible for reviewing it at the end of the process. That to me is the, is the kind of template that I just think goes across knowledge, work and odd. Cowork is a great example. Open Closet's a great example. You can kind of, sort of see what Codex could become over time. These are some, some really interesting kind of platforms that are emerging.swyx: Okay. Um, I wanted to, we touched on evals a little bit. You had, you had the report that you're gonna go bring up and then I was gonna go into like, uh, boxes, evals, but uh, go ahead. Talk about your genetic search thing.Jeff Huber: Yeah. Mostly I think kinda a few of the insights. It's like number one frontier model is not good at search.Humans have this [00:30:00] natural explore, exploit trade off where we kinda understand like when to stop doing something. Also, humans are pretty good at like forgetting actually, and like pruning their own context, whereas agents are not, and actually an agent in their kind of context history, if they knew something was bad and they even, you could see in the trace the reason you trace, Hey, that probably wasn't a good idea.If it's still in the trace, still in the context, they'll still do it again. Uhhuh. Uh, and so like, I think pruning is also gonna be like, really, it's already becoming a thing, right? But like, letting self prune the con windowsswyx: be a big deal. Yeah. So, so don't leave the mistake. Don't leave the mistake in there.Cut out the mistake but tell it that you made a mistake in the past and so it doesn't repeat it.Jeff Huber: Yeah. But like cut it out so it doesn't get like distracted by it again. ‘cause really, you know, what is so, so it will repeat its mistake just because it's been, it's inswyx: theJeff Huber: context. It'sAaron Levie: in the context so much.That's a few shot example. Even if it, yeah.Jeff Huber: It's like oh thisAaron Levie: is a great thing to go try even ifJeff Huber: it didn't work.Aaron Levie: Yeah,Jeff Huber: exactly.Aaron Levie: SoJeff Huber: there's like a bunch of stuff there. JustAaron Levie: Groundhogs Day inside these models. Yeah. I'm gonna go keep doing the same wrongJeff Huber: thing. Covering sense. I feel like, you know, some creator analogy you're trying like fit a manifold in latent space, which kind is doing break program synthesis, which is kinda one we think about we're doing right.Like, you know, certain [00:31:00] facts might be like sort of overly pitting it. There are certain, you know, sec sectors of latent space and so like plug clean space. Yeah. And, uh, andswyx: so we have a bell, our editor as a bell every time you say that. SoJeff Huber: you have, you have to like remove those, likeswyx: you shoulda a gong like TPN or something.IfJeff Huber: we gong, you either remove those links to like kinda give it the freedom, kind of do what you need to do. So, but yeah. We'll, we'll release more soon. That'sAaron Levie: awesome.Jeff Huber: That'll, that'll be cool.swyx: We're a cerebral podcast that people listen to us and, and sort of think really deep. So yeah, we try to keep it subtle.Okay. We try to keep it.Aaron Levie: Okay, fine.Inside Agent Evalsswyx: Um, you, you guys do, you guys do have EVs, you talked about your, your office thing, but, uh, you've been also promoting APEX agents and complex work. Uh, yeah, whatever you, wherever you wanna take this just Yeah. How youAaron Levie: Apex is, is obviously me, core's, uh, uh, kind of, um, agent eval.We, we supported that by sort of. Opening up some data for them around how we kind of see these, um, data workspaces in, in the, you know, kind of regular economy. So how do lawyers have a workspace? How do investment bankers have a workspace? What kind of data goes into those? And so we, [00:32:00] we partner with them on their, their apex eval.Our own, um, eval is, it's actually relatively straightforward. We have a, a set of, of documents in a, in a range of industries. We give the agent previously did this as a one shot test of just purely the model. And then we just realized we, we need to, based on where everything's going, it's just gotta be more agentic.So now it's a bit more of a test of both our harness and the model. And we have a rubric of a set of things that has to get right and we score it. Um, and you're just seeing, you know, these incredible jumps in almost every single model in its own family of, you know, opus four, um, you know, sonnet four six versus sonnet four five.swyx: Yeah. We have this up on screen.Aaron Levie: Okay, cool. So some, you're seeing it somewhere like. I, I forget the to, it was like 15 point jump, I think on the main, on the overall,swyx: yes.Aaron Levie: And it's just like, you know, these incredible leaps that, that are starting to happen. Um,swyx: and OP doesn't know any, like any, it's completely held out from op.Aaron Levie: This is not in any, there's no public data which has, you know, Ben benefits and this is just a private eval that we [00:33:00] do, and then we just happen to show it to, to the world. Hmm. So you can't, you can't train against it. And I think it's just as representative of. It's obviously reasoning capabilities, what it's doing at, at, you know, kind of test time, compute capabilities, thinking levels, all like the context rot issues.So many interesting, you know, kind of, uh, uh, capabilities that are, that are now improvingswyx: one sector that you have. That's interesting.Industries and Datasetsswyx: Uh, people are roughly familiar with healthcare and legal, but you have public sector in there.Aaron Levie: Yeah.swyx: Uh, what's that? Like, what, what, what is that?Aaron Levie: Yeah, and, and we actually test against, I dunno, maybe 10 industries.We, we end up usually just cutting a few that we think have interesting gains. All extras, won a lot of like government type documents. Um,swyx: what is that? What is it? Government type documents?Aaron Levie: Government filings. Like a taxswyx: return, likeAaron Levie: a probably not tax returns. It would be more of what would go the government be using, uh, as data.So, okay. Um, so think about research that, that type of, of, of data sets. And then we have financial services for things like data rooms and what would be in an investment prospectus. Uhhuh,swyx: that one you can dog food.Aaron Levie: Yeah, exactly. Exactly. Yes. Yes. [00:34:00] So, uh, so we, we run the models, um, in now, you know, more of an agent mode, but, but still with, with kinda limited capacity and just try and see like on a, like, for like basis, what are the improvements?And, and again, we just continue to be blown away by. How, how good these models are getting.swyx: Yeah, I mean, I think every serious AI company needs something like that where like, well, this is the work we do. Here's our company eval. Yeah. And if you don't have it, well, you're not a serious AI company.Aaron Levie: There's two dimensions, right?So there's, there's like, how are the models improving? And so which models should you either recommend a customer use, which one should you adopt? But then every single day, we're making changes to our agents. And you need to knowswyx: if you regressed,Aaron Levie: if you know. Yeah. You know, I've been fully convinced that the whole agent observability and eval space is gonna be a massive space.Um, super excited for what Braintrust is doing, excited for, you know, Lang Smith, all the things. And I think what you're going to, I mean, this is like every enter like literally every enterprise right now. It's like the AI companies are the customers of these tools. Every enterprise will have this. Yeah, you'll just [00:35:00] have to have an eval.Of all of your work and like, we'll, you'll have an eval of your RFP generation, you'll have an eval of your sales material creation. You'll have an eval of your, uh, invoice processing. And, and as you, you know, buy or use new agentic systems, you are gonna need to know like, what's the quality of your, of your pipeline.swyx: Yeah.Aaron Levie: Um, so huge, huge market with agent evals.swyx: Yeah.Building the Agent Teamswyx: And, and you know, I'm gonna shout out your, your team a bit, uh, your CTO, Ben, uh, did a great talk with us last year. Awesome. And he's gonna come back again. Oh, cool. For World's Fair.Aaron Levie: Yep.swyx: Just talk about your team, like brag a little bit. I think I, I think people take these eval numbers in pretty charts for granted, but No, there, I mean, there's, there's lots of really smart people at work during all this.Aaron Levie: Biggest shout out, uh, is we have a, we have a couple folks at Dya, uh, Sidarth, uh, that, that kind of run this. They're like a, you know, kind of tag tag team duo on our evals, Ben, our CTO, heavily involved Yasha, head of ai, uh, you know, a bunch of folks. And, um, evals is one part of the story. And then just like the full, you know, kind of AI.An agent team [00:36:00] is, uh, is a, is a pretty, you know, is core to this whole effort. So there's probably, I don't know, like maybe a few dozen people that are like the epicenter. And then you just have like layers and layers of, of kind of concentric circles of okay, then there's a search team that supports them and an infrastructure team that supports them.And it's starting to ripple through the entire company. But there's that kind of core agent team, um, that's a pretty, pretty close, uh, close knit group.swyx: The search team is separate from the infra team.Aaron Levie: I mean, we have like every, every layer of the stack we have to kind of do, except for just pure public cloud.Um, but um, you know, we, we store, I don't even know what our public numbers are in, you know, but like, you can just think about it as like a lot of data is, is stored in box. And so we have, and you have every layer of the, of the stack of, you know, how do you manage the data, the file system, the metadata system, the search system, just all of those components.And then they all are having to understand that now you've got this new customer. Which is the agent, and they've been building for two types of customers in the past. They've been building for users and they've been building for like applications. [00:37:00] And now you've got this new agent user, and it comes in with a difference of it, of property sometimes, like, hey, maybe sometimes we should do embeddings, an embedding based, you know, kind of search versus, you know, your, your typical semantic search.Like, it's just like you have to build the, the capabilities to support all of this. And we're testing stuff, throwing things away, something doesn't work and, and not relevant. It's like just, you know, total chaos. But all of those teams are supporting the agent team that is kind of coming up with its requirements of what, what do we need?swyx: Yeah. No, uh, we just came from, uh, fireside chat where you did, and you, you talked about how you're doing this. It's, it's kind of like an internal startup. Yeah. Within the broader company. The broader company's like 3000 people. Yeah. But you know, there's, there's a, this is a core team of like, well, here's the innovation center.Aaron Levie: Yeah.swyx: And like that every company kind of is run this way.Aaron Levie: Yeah. I wanna be sensitive. I don't call it the innovation center. Yeah. Only because I think everybody has to do innovation. Um, there, there's a part of the, the, the company that is, is sort of do or die for the agent wave.swyx: Yeah.Aaron Levie: And it only happens to be more of my focus simply because it's existential that [00:38:00] we get it right.swyx: Yeah.Aaron Levie: All of the supporting systems are necessary. All of the surrounding adjacent capabilities are necessary. Like the only reason we get to be a platform where you'd run an agent is because we have a security feature or a compliance feature, or a governance feature that, that some team is working on.But that's not gonna be the make or break of, of whether we get agents right. Like that already exists and we need to keep innovating there. I don't know what the right, exact precise number is, but it's not a thousand people and it's not 10 people. There's a number of people that are like the, the kind of like, you know, startup within the company that are the make or break on everything related to AI agents, you know, leveraging our platform and letting you work with your data.And that's where I spend a lot of my time, and Ben and Yosh and Diego and Teri, you know, these are just, you know, people that, that, you know, kind of across the team. Are working.swyx: Yeah. Amazing.Read Write Agent WorkflowsJeff Huber: How do you, how do you think about, I mean, you talked a lot about like kinda read workflows over your box data. Yep.Right. You know, gen search questions, queries, et cetera. But like, what about like, write or like authoring workflows?Aaron Levie: Yes. I've [00:39:00] already probably revealed too much actually now that I think about it. So, um, I've talked about whatever,Jeff Huber: whatever you can.Aaron Levie: Okay. It's just us. It's just us. Yeah. Okay. Of course, of course.So I, I guess I would just, uh, I'll make it a little bit conceptual, uh, because again, I've already, I've already said things that are not even ga but, but we've, we've kinda like danced around it publicly, so I, yeah, yeah. Okay. Just like, hopefully nobody watches this, um, episode. No.swyx: It's tidbits for the Heidi engaged to go figure out like what exactly, um, you know, is, is your sort of line of thinking.Sure. They can connect the dots.Aaron Levie: Yeah. So, so I would say that, that, uh, we, you know, as a, as a place where you have your enterprise content, there's a use case where I want to, you know, have an agent read that data and answer questions for me. And then there's a use case where I want the agent to create something.And use the file system to create something or store off data that it's working on, or be able to have, you know, various files that it's writing to about the work it's doing. So we do see it as a total read write. The harder problem has so far been the read only because, because again, you have that kind of like 10 [00:40:00] million to one ratio problem, whereas rights are a lot of, that's just gonna come from the model and, and we just like, we'll just put it in the file system and kinda use it.So it's a little bit of a technically easier problem, but the only part that's like, not necessarily technically hard, it is just like it's not yet perfected in the state of the ecosystem is, you know, building a beautiful PowerPoint presentation. It's still a hard problem for these models. Like, like we still, you know, like, like these formats are just, we're not built for.They'reswyx: working on it.Aaron Levie: They're, they're working on it. Everybody's working on it.swyx: Every launch is like, well, we do PowerPoint now.Aaron Levie: We're getting, yeah, getting a lot, getting a lot of better each time. But then you'll do this thing where you'll ask the update one slide and all of a sudden, like the fonts will be just like a little bit different, you know, on two of the slides, or it moved, you know, some shape over to the left a little bit.And again, these are the kind of things that, like in code, obviously you could really care about if you really care about, you know, how beautiful is the code, but at the end, user doesn't notice all those problems and file creation, the end user instantly sees it. You're [00:41:00] like, ah, like paragraph three, like, you literally just changed the font on me.Like it's a totally different font and like midway through the document. Mm-hmm. Those are the kind of things that you run into a lot of in the, in the content creation side. So, mm-hmm. We are gonna have native agents. That do all of those things, they'll be powered by the leading kind of models and labs.But the thing that I think is, is probably gonna be a much bigger idea over time is any agent on any system, again, using Box as a file system for its work, and in that kind of scenario, we don't necessarily care what it's putting in the file system. It could put its memory files, it could put its, you know, specification, you know, documents.It could put, you know, whatever its markdown files are, or it could, you know, generate PDFs. It's just like, it's a workspace that is, is sort of sandboxed off for its work. People can collaborate into it, it can share with other people. And, and so we, we were thinking a lot about what's the right, you know, kind of way to, to deliver that at scale.Docs Graphs and Founder Modeswyx: I wanted to come into sort of the sort of AI transformation or AI sort of, uh, operations things. [00:42:00] Um, one of the tweets that you, that you wanted to talk about, this is just me going through your tweets, by the way. Oh, okay. I mean, like, this is, you readAaron Levie: one by one,swyx: you're the, you're the easiest guest to prep for because you, you already have like, this is the, this is what I'm interested in.I'm like, okay, well, areAaron Levie: we gonna get to like, like February, January or something? Where are we in the, in the timelines? How far back are we going?swyx: Can you, can you describe boxes? A set of skills? Right? Like that, that's like, that's like one of the extremes of like, well if you, you just turn everything into a markdown file.Yeah. Then your agent can run your company. Uh, like you just have to write, find the right sequence of words toAaron Levie: Yes.swyx: To do it.Aaron Levie: Sorry, isthatswyx: the question? So I think the question is like, what if we documented everything? Yes. The way that you exactly said like,Aaron Levie: yes.swyx: Um, let's get all the Fortune five hundreds, uh, prepared for agents.Yes. And like, you know, everything's in golden and, and nicely filed away and everything. Yes. What's missing? Like, what's left, right? LikeAaron Levie: Yeah.swyx: You've, you've run your company for a decade. LikeAaron Levie: Yeah. I think the challenge is that, that that information changes a week later. And because something happened in the market for that [00:43:00] customer, or us as a company that now has to go get updated, and so these systems are living and breathing and they have to experience reality and updates to reality, which right now is probably gonna be humans, you know, kinda giving those, giving them the updates.And, you know, there is this piece about context graphs as as, uh, that kinda went very viral. Yeah. And I, I, I was like a, i, I, I thought it was super provocative. I agreed with many parts of it. I disagree with a few parts around. You know, it's not gonna be as easy as as just if we just had the agent traces, then we can finally do that work because there's just like, there's so much more other stuff that that's happening that, that we haven't been able to capture and digitize.And I think they actually represented that in the piece to be clear. But like there's just a lot of work, you know, that that has to, you just can't have only skills files, you know, for your company because it's just gonna be like, there's gonna be a lot of other stuff that happens. Yeah. Change over time.Yeah. Most companies are practically apprenticeships.swyx: Most companies are practically apprenticeships. LikeJeff Huber: every new employee who joins the team, [00:44:00] like you span one to three months. Like ramping them up.Aaron Levie: Yes. AllJeff Huber: that tat knowledgeAaron Levie: isJeff Huber: not written down.Aaron Levie: Yes.Jeff Huber: But like, it would have to be if you wanted to like give it to an Asian.Right. And so like that seems to me like to beAaron Levie: one is I think you're gonna see again a premium on companies that can document this. Mm-hmm. Much. There'll be a huge premium on that because, because you know, can you shorten that three month ramp cycle to a two week ramp cycle? That's an instant productivity gain.Can you re dramatically reduce rework in the organization because you've documented where all the stuff is and where the answers are. Can you make your average employee as good as your 90th percentile employee because you've captured the knowledge that's sort of in the heads of, of those top employees and make that available.So like you can see some very clear productivity benefits. Mm-hmm. If you had a company culture of making sure you know your information was captured, digitized, put in a format that was agent ready and then made available to agents to work with, and then you just, again, have this reality of like add a 10,000 person [00:45:00] company.Mapping that to the, you know, access structure of the company is just a hard problem. Is like, is like, yeah, well, you just, not every piece of information that's digitized can be shared to everybody. And so now you have to organize that in a way that actually works. There was a pretty good piece, um, this, this, uh, this piece called your company as a file is a file system.I, did you see that one?swyx: Nope.Aaron Levie: Uh, yes. You saw it. Yeah. And, and, uh, I actually be curious your thoughts on it. Um, like, like an interesting kind of like, we, we agree with it because, because that's how we see the world and, uh,swyx: okay. We, we have it up on screen. Oh,Aaron Levie: okay. Yeah. But, but it's all about basically like, you know, we've already, we, we, we already organized in this kind of like, you know, permission structure way.Uh, and, and these are the kind of, you know, natural ways that, that agents can now work with data. So it's kind of like this, this, you know, kind of interesting metaphor, but I do think companies will have to start to think about how they start to digitize more, more of that data. What was your take?Jeff Huber: Yeah, I mean, like the company's probably like an acid compliant file system.Aaron Levie: Uh,Jeff Huber: yeah. Which I'm guessing boxes, right? So, yeah. Yes.swyx: Yeah. [00:46:00]Jeff Huber: Which you have a great piece on, but,swyx: uh, yeah. Well, uh, I, I, my, my, my direction is a little bit like, I wanna rewind a little bit to the graph word you said that there, that's a magic trigger word for us. I always ask what's your take on knowledge graphs?Yeah. Uh, ‘cause every, especially at every data database person, I just wanna see what they think. There's been knowledge graphs, hype cycles, and you've seen it all. So.Aaron Levie: Hmm. I actually am not the expert in knowledge graphs, so, so that you might need toswyx: research, you don't need to be an expert. Yeah. I think it's just like, well, how, how seriously do people take it?Yeah. Like, is is, is there a lot of potential in the, in the HOVI?Aaron Levie: Uh, well, can I, can I, uh, understand first if it's, um, is this a loaded question in the sense of are you super pro, super con, super anti medium? Iswyx: see pro, I see pros and cons. Okay. Uh, but I, I think your opinion should be independent of mine.Aaron Levie: Yeah. No, no, totally. Yeah. I just want to see what I'm stepping into.swyx: No, I know. It's a, and it's a huge trigger word for a lot of people out Yeah. In our audience. And they're, they're trying to figure out why is that? Because whyAaron Levie: is this such aswyx: hot item for them? Because a lot of people get graph religion.And they're like, everything's a graph. Of course you have to represent it as a graph. Well, [00:47:00] how do you solve your knowledge? Um, changing over time? Well, it's a graph.Aaron Levie: Yeah.swyx: And, and I think there, there's that line of work and then there's, there's a lot of people who are like, well, you don't need it. And both are right.Aaron Levie: Yeah. And what do the people who say you don't need it, what are theyswyx: arguing for Mark down files. Oh, sure, sure. Simplicity.Aaron Levie: Yeah.swyx: Versus it's, it's structure versus less structure. Right. That's, that's all what it is. I do.Aaron Levie: I think the tricky thing is, um, is, is again, when this gets met with real humans, they're just going to their computer.They're just working with some people on Slack or teams. They're just sharing some data through a collaborative file system and Google Docs or Box or whatever. I certainly like the vision of most, most knowledge graph, you know, kind of futuristic kind of ways of thinking about it. Uh, it's just like, you know, it's 2026.We haven't seen it yet. Kind of play out as as, I mean, I remember. Do you remember the, um, in like, actually I don't, I don't even know how old you guys are, but I'll for, for to show my age. I remember 17 years ago, everybody thought enterprises would just run on [00:48:00] Wikis. Yeah. And, uh, confluence and, and not even, I mean, confluence actually took off for engineering for sure.Like unquestionably. But like, this was like everything would be in the w. And I think based on our, uh, our, uh, general style of, of, of what we were building, like we were just like, I don't know, people just like wanna workspace. They're gonna collaborate with other people.swyx: Exactly. Yeah. So you were, you were anti-knowledge graph.Aaron Levie: Not anti, not anti. Soswyx: not nonAaron Levie: I'm not, I'm not anti. ‘cause I think, I think your search system, I just think these are two systems that probably, but like, I'm, I'm not in any religious war. I don't want to be in anybody's YouTube comments on this. There's not a fight for me.swyx: We, we love YouTube comments. We're, we're, we're get into comments.Aaron Levie: Okay. Uh, but like, but I, I, it's mostly just a virtue of what we built. Yeah. And we just continued down that path. Yeah.swyx: Yeah.Aaron Levie: And, um, and that, that was what we pursued. But I'm not, this is not a, you know, kind of, this is not a, uh, it'sswyx: not existential for you. Great.Aaron Levie: We're happy to plug into somebody else's graph.We're happy to feed data into it. We're happy for [00:49:00] agents to, to talk to multiple systems. Not, not our fight.swyx: Yeah.Aaron Levie: But I need your answer. Yeah. Graphs or nerd Snipes is very effective nerd.swyx: See this is, this is one, one opinion and then I've,Jeff Huber: and I think that the actual graph structure is emergent in the mind of the agent.Ah, in the same way it is in the mind of the human. And that's a more powerful graph ‘cause it actually involved over time.swyx: So don't tell me how to graph. I'll, I'll figure it out myself. Exactly. Okay. All right. AndJeff Huber: what's yours?swyx: I like the, the Wiki approach. Uh, my, I'm actually

As the new generation of financial infrastructure takes shape, the need for sophisticated tools to inform policy-makers is growing. To encourage innovation and preserve security, it is vital that policy-makers have the capacity to interact with and closely supervise new systems. Sandboxes are a key tool for this process. Carmelle Cadet, chief executive officer and founder of EmTech joins OMFIF to discuss the nuances of how sandboxes work and how they should be constructed.

Regulatory sandboxes allow startups to experiment with AI technologies in a controlled setting, fostering innovation while maintaining oversight. Joseph Joshy from IFSCA highlights their importance at the India AI Impact Summit 2026, noting they help balance innovation with risk management. Industry leaders discuss the need for India-specific AI protocols and emphasize the role of digital infrastructure in AI advancement.Learn more on this news by visiting us at: https://greyjournal.net/news/ Hosted on Acast. See acast.com/privacy for more information.

Kah Kit Yip has spent his career turning bold ideas about the future of money into working infrastructure.After nearly two decades at Malaysia's central bank and leading breakthrough initiatives at the Bank for International Settlements, he helped pioneer new models for instant cross-border payments, programmable compliance, and faster, safer FX settlement. Now, as Head of Blockchain and Digital Assets at UOB, he's bringing those concepts out of policy sandboxes and into production across Southeast Asia.In this conversation, Ari and Kah Kit explore how interoperable payment rails, wholesale CBDCs, and tokenized assets can move value across borders in real time — while meeting the highest standards for compliance and risk management. They discuss why legal clarity and shared standards are essential for adoption, how institutions can match the speed of digital payments with equally fast safeguards, and what it takes for regulators and industry to build global financial infrastructure together.Along the way, Kah Kit reflects on marathon running, crime novels, and the grit required to solve complex, long-horizon challenges in finance and technology.

REGULATING ARTIFICIAL INTELLIGENCE Colleague Kevin Frazier. Kevin Frazier continues, warning against a "waterfall of regulation" by states and advocating for "regulatory sandboxes" to allow experimentation. NUMBER 8 NOVEMBER 1955

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos? The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews.  You can find all the music from the segment about encores in this YouTube playlist. This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer. If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281. Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Learn more about your ad choices. Visit megaphone.fm/adchoices

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos?The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews. You can find all the music from the segment about encores in this YouTube playlist.This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer.If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281.Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Hosted on Acast. See acast.com/privacy for more information.

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos? The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews.  You can find all the music from the segment about encores in this YouTube playlist. This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer. If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281. Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Learn more about your ad choices. Visit megaphone.fm/adchoices

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos?The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews. You can find all the music from the segment about encores in this YouTube playlist.This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer.If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281.Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Hosted on Acast. See acast.com/privacy for more information.

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos?The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews. You can find all the music from the segment about encores in this YouTube playlist.This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer.If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281.Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Hosted on Acast. See acast.com/privacy for more information.

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos? The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews.  You can find all the music from the segment about encores in this YouTube playlist. This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer. If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281. Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Learn more about your ad choices. Visit megaphone.fm/adchoices

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos?The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews. You can find all the music from the segment about encores in this YouTube playlist.This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer.If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281.Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Hosted on Acast. See acast.com/privacy for more information.

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos? The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews.  You can find all the music from the segment about encores in this YouTube playlist. This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer. If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281. Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Learn more about your ad choices. Visit megaphone.fm/adchoices

Decoder Ring listeners write in with some excellent mysteries, and for our last episode of the year we're solving three of them. Why do children play in boxes full of sand? Why do rock bands pretend like the show is over when everybody knows they're coming back for an encore? And what was up with those school assemblies where you'd get to skip class to learn about…yo-yos?The voices you'll hear in this episode include yo-yo masters ”Dazzling Dave” Schulte and Dale Oliver, children's book author Rob Peñas, Pulitzer Prize-winning design critic Alexandra Lange, and music journalists Brian Wise, Michael Walker, and Travis Andrews. You can find all the music from the segment about encores in this YouTube playlist.This episode was produced by Max Freedman, Katie Shepherd, and Evan Chung, Decoder Ring's supervising producer. Merritt Jacob is Senior Technical Director. We had additional production from Joel Meyer.If you have any cultural mysteries you want us to decode, email us at DecoderRing@slate.com or leave a message on our hotline at (347) 460-7281.Get more of Decoder Ring with Slate Plus! Join for exclusive bonus episodes of Decoder Ring and ad-free listening on all your favorite Slate podcasts. Subscribe from the Decoder Ring show page on Apple Podcasts or Spotify. Or, visit slate.com/decoderplus for access wherever you listen. Hosted on Acast. See acast.com/privacy for more information.

Nach einer kurzen Herbstpause sprechen Alois und Oliver darüber, wie echte Deep-Tech-Ökosysteme auf Stadtebene entstehen – jenseits von Buzzwords. Sie ordnen ein, was Deep Tech wirklich bedeutet, wo die Grenze zu Commodity-Technologien verläuft und weshalb der Begriff aktuell oft inflationär verwendet wird. Ausgehend von Deutschlands neuer Hightech-Agenda (mit Schwerpunkten wie KI, Quantencomputing und Biotechnologie) diskutieren sie, wie nationale Programme lokale Cluster beeinflussen und warum Orte mit starker Forschungs- und Infrastrukturbasis – etwa DESY und die Science City in Hamburg – als Anker dienen.Die Episode zeigt, wie Konvergenz zwischen Technologien wie KI, Quanten- und Neuromorphik-Computing, Robotik oder XR Tempo aufnimmt und welche Rolle „universelle Schnittstellen“ wie große Sprachmodelle dabei spielen. Zugleich machen die beiden deutlich, dass Grundlagenforschung geschützte Domänen braucht, während Reallabore und Sandboxes die Brücke in die Anwendung schlagen. An konkreten Beispielen wird sichtbar, weshalb kurze Wege, stabile Rahmenbedingungen, verlässliche Testflächen und klar profilierte Stadt-Stärken den Unterschied zwischen Theorie und Transfer ausmachen.Ein zentrales Motiv der Folge ist der „Übersetzer“: Personen und Formate, die Forschung, Industrie und – oft als Ankerkunden – den Staat zielgerichtet zusammenbringen. Statt Event-Inflation braucht es kuratierte Räume, in denen Themen kurz vor der Marktreife fokussiert beschleunigt werden. Genau hier setzt der Deep Tech Campus Circle an: als bewusst exklusiver Ort, an dem Köpfe, Kapital und konkrete Use Cases aufeinandertreffen, um aus Keimlingen skalierbare Innovationen zu machen.Am Ende bleibt die klare These: Deep Tech entsteht, wenn tiefe Expertise, Geduld und ein klug orchestriertes Ökosystem zusammenwirken. Wer verstehen will, wie Städte vom Strategiepapier zur sichtbaren Wertschöpfung kommen, bekommt in dieser Folge eine kompakte, praxisnahe Landkarte – von der Saat in der Grundlagenforschung bis zur Ernte im Markt.

Vasek Mlejnsky is Co-Founder & CEO of E2B, the open-source runtime for executing AI-generated code in secure cloud sandboxes. Essentially, they give AI agents cloud computers. Their open source repos, particularly e2b which has 9K GitHub stars, have been widely adopted to help securely run AI-generated code. E2B has raised $12M from investors including Decibel and Sunflower. In this episode, we dig into:Why agents need a sandboxBuilding a new category of infra tooling, much like LaunchDarkly Some of their viral content moments - including Greg Brockman sharing their videosFiguring out the right commercial offering Why they don't agree with pricing per token Why moving from Prague to the Bay Area felt essential for them as founders

The project, spearheaded by Greek non-profit child welfare organisation The Smile of the Child, will be co-created by children and young people to ensure their voices are heard. The ISPCC is honoured to announce its participation in a worldwide project designed to transform how we prevent and respond to online child sexual exploitation and abuse. Safe Online, a global fund dedicated to eradicating online child sexual exploitation and abuse, is funding the project called "Sandboxing and Standardising Child Online Redress". The COR Sandbox project will establish a first-of-its-kind mechanism to advance child online safety through collaboration across sectors, borders and generations. The project is led by The Smile of the Child, Greece's premier child welfare organisation and ISPCC is a partner alongside The Young and Resilient Research Centre at Western Sydney University, Child Helpline International and the Centre for Digital Policy at University College Dublin. Sandboxes bring together industry, regulators and customers in a safe space to test innovative products and services without incurring regulatory sanctions and they are mainly used in the finance sector to test new services. The EU is increasingly encouraging the use of sandboxes in the field of high technology and artificial intelligence. Through the participation of youth, platforms, regulators and online safety experts, this first regulatory sandbox for child digital wellbeing will provide for consistent, systemic care and redress for children from online harm, based on their rights under the United Nations Convention on the Rights of the Child (UNCRC). Getting reporting and redress right means that we can keep track of harms and be able to identify systemic risk. Co-designing the reporting and redress process with young people as equitable participants can help us understand what they expect from the reporting process and what remedies are fair for them putting Article 12 of the UNCRC into action. The project also benefits from the guidance of renowned digital safety experts, including Project Lead and Scientific Coordinator Ioanna Noula, PhD, an international expert on tech policy and children's rights; pioneering online safety and youth rights advocate Anne Collier; youth rights and participation expert Amanda Third, PhD, of the Young and Resilient Research Centre; international innovation management consultant Nicky Hickman; IT innovation and startup founder Jez Goldstone; and leading child online wellbeing scholar Tijana Milosevic, PhD. ISPCC Head of Policy and Public Affairs Fiona Jennings said: "This project is a wonderful example of what we can achieve when we collaborate and listen to children and young people. Having robust online reporting mechanisms in place is a key policy objective for ISPCC and this project will go a long way towards making the online world safer for children and young people to participate in." Project lead Ioanna Noula said: "ISPCC's contribution to a project, which seeks to build coherence around the issue of online redress, will be a catalyst for real and substantial change in the area of online reporting. Helplines play a key role in flagging illegal and/or harmful content. As the experts in listening and responding to children, ISPCC can provide insight from an Irish context to help spearheading the implementation of the Digital Services Act and the wellbeing of children online." See more stories here. More about Irish Tech News Irish Tech News are Ireland's No. 1 Online Tech Publication and often Ireland's No.1 Tech Podcast too. You can find hundreds of fantastic previous episodes and subscribe using whatever platform you like via our Anchor.fm page here: https://anchor.fm/irish-tech-news If you'd like to be featured in an upcoming Podcast email us at Simon@IrishTechNews.ie now to discuss. Irish Tech News have a range of services available to help promote your business. Why not drop us a line at Info@IrishTechNews.ie now to ...

Vasek Mlejnsky from E2B joins us today to talk about sandboxes for AI agents. In the last 2 years, E2B has grown from a handful of developers building on it to being used by ~50% of the Fortune 500 and generating millions of sandboxes each week for their customers. As the “death of chat completions” approaches, LLMs workflows and agents are relying more and more on tool usage and multi-modality.The most common use cases for their sandboxes:- Run data analysis and charting (like Perplexity)- Execute arbitrary code generated by the model (like Manus does)- Running evals on code generation (see LMArena Web)- Doing reinforcement learning for code capabilities (like HuggingFace)Full Video EpisodeTimestamps00:00:00 Introductions00:00:37 Origin of DevBook -> E2B00:02:35 Early Experiments with GPT-3.5 and Building AI Agents00:05:19 Building an Agent Cloud00:07:27 Challenges of Building with Early LLMs00:10:35 E2B Use Cases00:13:52 E2B Growth vs Models Capabilities00:15:03 The LLM Operating System (LLMOS) Landscape00:20:12 Breakdown of JavaScript vs Python Usage on E2B00:21:50 AI VMs vs Traditional Cloud00:26:28 Technical Specifications of E2B Sandboxes00:29:43 Usage-based billing infrastructure00:34:08 Pricing AI on Value Delivered vs Token Usage00:36:24 Forking, Checkpoints, and Parallel Execution in Sandboxes00:39:18 Future Plans for Toolkit and Higher-Level Agent Frameworks00:42:35 Limitations of Chat-Based Interfaces and the Future of Agents00:44:00 MCPs and Remote Agent Capabilities00:49:22 LLMs.txt, scrapers, and bad AI bots00:53:00 Manus and Computer Use on E2B00:55:03 E2B for RL with Hugging Face00:56:58 E2B for Agent Evaluation on LMArena00:58:12 Long-Term Vision: E2B as Full Lifecycle Infrastructure for LLMs01:00:45 Future Plans for Hosting and Deployment of LLM-Generated Apps01:01:15 Why E2B Moved to San Francisco01:05:49 Open Roles and Hiring Plans at E2B Get full access to Latent.Space at www.latent.space/subscribe

Vasek Mlejnsky from E2B joins us today to talk about sandboxes for AI agents. In the last 2 years, E2B has grown from a handful of developers building on it to being used by ~50% of the Fortune 500 and generating millions of sandboxes each week for their customers. As the “death of chat completions” approaches, LLMs workflows and agents are relying more and more on tool usage and multi-modality. The most common use cases for their sandboxes: - Run data analysis and charting (like Perplexity) - Execute arbitrary code generated by the model (like Manus does) - Running evals on code generation (see LMArena Web) - Doing reinforcement learning for code capabilities (like HuggingFace) Timestamps: 00:00:00 Introductions 00:00:37 Origin of DevBook -> E2B 00:02:35 Early Experiments with GPT-3.5 and Building AI Agents 00:05:19 Building an Agent Cloud 00:07:27 Challenges of Building with Early LLMs 00:10:35 E2B Use Cases 00:13:52 E2B Growth vs Models Capabilities 00:15:03 The LLM Operating System (LLMOS) Landscape 00:20:12 Breakdown of JavaScript vs Python Usage on E2B 00:21:50 AI VMs vs Traditional Cloud 00:26:28 Technical Specifications of E2B Sandboxes 00:29:43 Usage-based billing infrastructure 00:34:08 Pricing AI on Value Delivered vs Token Usage 00:36:24 Forking, Checkpoints, and Parallel Execution in Sandboxes 00:39:18 Future Plans for Toolkit and Higher-Level Agent Frameworks 00:42:35 Limitations of Chat-Based Interfaces and the Future of Agents 00:44:00 MCPs and Remote Agent Capabilities 00:49:22 LLMs.txt, scrapers, and bad AI bots 00:53:00 Manus and Computer Use on E2B 00:55:03 E2B for RL with Hugging Face 00:56:58 E2B for Agent Evaluation on LMArena 00:58:12 Long-Term Vision: E2B as Full Lifecycle Infrastructure for LLMs 01:00:45 Future Plans for Hosting and Deployment of LLM-Generated Apps 01:01:15 Why E2B Moved to San Francisco 01:05:49 Open Roles and Hiring Plans at E2B

In this episode of The Stream Life Podcast, Joel Vincent joins the show to talk about Cribl's latest innovation: a Lakehouse that's purpose built for telemetry data.  Resources Read the blog If you want to automatically get every episode of the Stream Life podcast, you can subscribe on your favorite podcast app. Cribl, the Data Engine for IT and Security, empowers organizations to transform their data strategy. Customers use Cribl's suite of products to collect, process, route, and analyze all IT and security data, delivering the flexibility, choice, and control required to adapt to their ever-changing needs. We offer free training, certifications, and a free tier across our products. Our community Slack features Cribl engineers, partners, and customers who can answer your questions as you get started and continue to build and evolve. We also offer a variety of hands-on Sandboxes for those interested in how companies globally leverage our products for their data challenges.

KFI White House correspondent Jon Decker speaks on the latest out of the Oval Office. Google joins the list of companies distancing themselves from DEI. California EV sales flat in 2024. Can Newsom's mandate be met? Researchers tested sandboxes and street dust for lead after Eaton Fire… Here's what they found.

Have you ever left a sales meeting feeling uncertain whether your potential client will buy? If you've experienced clients nodding along but still feeling overwhelmed, you're not alone. In this solo episode, I share two powerful actions you can take to simplify the buying process for your clients. First, we'll explore the importance of using a visual industry map that highlights their journey and pain points. Then, I'll discuss how building an industry-specific sandbox allows clients to interact with your solution tangibly. These strategies have proven effective for tech consultants and partners across various industries, and I'm excited to help you implement them in your own sales process.Resources and LinksPrevious episode: 589 - AI 1st business with John MunsellCheck out more episodes of The Paul Higgins ShowSubscribe to our YouTube channel: @PaulHigginsMentoringThe Tech Consultant's RoadmapJoin our newsletterJoin the Tech CollectiveSuggested resourcesFind out more about Paul and how he can help you

Imagine you're in a facebook group chat with a bunch of people and someone asks what constitutes a sandbox in a TTRPG. How much does improv vs. preparation factor into it? Can the players go off and do absolutely anything they want? What happens when you run a complete dungeon crawl? Let's talk about it! ---------------------------------------------------------------- Want to watch these episodes live? Check us out at https://www.youtube.com/@somederpsplaygames or twitch.tv/somederpsplaygames Check out the podcast on Soundcloud: https://soundcloud.com/somederpstalkaboutgames Want to tell us something? Email us at podcast@somederpsplaygames.com Like our Facebook page too! www.facebook.com/SomeDerpsPlayGames/ We have a Patreon! https://www.patreon.com/somederpsplaygames Rate us on iTunes! https://itunes.apple.com/us/podcast/some-derps-talk-about-games/id1048899720 Follow us on Twitter! SDPG: twitter.com/somederps Buddy: twitter.com/thatbuddysola Mango: twitter.com/theonetruemango Intro and Outro courtesy of twitter.com/VinceRolin