Podcasts about series b

In this episode, Jason Pritchard is joined by Austin Spiegel, CTO and Co-Founder of SIFT, to explore how better data infrastructure is accelerating advanced air mobility development. Austin explains how SIFT's hardware observability platform helps OEMs capture, organize and analyze vast volumes of telemetry generated across the entire aircraft lifecycle—from simulation and hardware-in-the-loop testing to manufacturing and flight operations. With certification programmes placing increasing pressure on traceability and auditability, he highlights how fragmented data systems, siloed workflows and manual reporting processes can create costly bottlenecks for eVTOL developers. The conversation dives into how SIFT creates a unified, structured “digital thread” that links telemetry to specific assets, sub-assemblies and manufacturing processes—enabling clear, regulator-ready evidence paths. Austin also discusses real-time flight test monitoring, automated pass/fail reporting, anomaly detection through “family” data comparisons, and the growing regulatory shift toward digital-ready, interactive evidence rather than static PDFs. With SIFT recently announcing a $42 million Series B raise, Austin shares how the company is scaling its team and capabilities to support the next phase of growth in aviation and beyond—building toward a single source of truth for hardware data that helps OEMs reduce late-stage surprises and move more confidently toward certification.

Synopsis: At the heart of JPM 2026's biotech buzz, Alok Tayi sits down with Fred Aslan, CEO of Artiva, to explore how bold platform bets, scalable cell therapies, and autoimmune breakthroughs could reshape medicine. Fred traces his journey from medical school in Brazil to consulting at BCG, venture capital, and ultimately founding multiple companies—sharing why following curiosity, not rigid career ladders, shaped his path. Fred dives deep into the bottlenecks holding back traditional CAR-T therapies—manufacturing complexity, cost, hospitalization, and toxicity—and explains how Artiva's off-the-shelf NK-cell platform aims to change the paradigm. The discussion explores why rheumatoid arthritis became Artiva's lead indication, how immune “resets” could redefine autoimmune care, and what's ahead in 2026 as the company prepares registrational trials and expands its basket studies across lupus, myositis, scleroderma, and more. The episode closes with rapid-fire takes on AI in drug development, China's accelerating biotech engine, rare disease trial models, and the strategic principles founders should follow when choosing indications and building durable platforms. Biography: Fred Aslan, M.D., has a 20-year track record as an executive and investor in the life sciences industry. He was most recently President and CBO at Vividion Therapeutics, where he was responsible for business development, finance, alliance and project management, and operations. Dr. Aslan had the opportunity to lead Vividion's Series B financing and $135M-upfront collaboration with Roche. Prior to Vividion, Dr. Aslan had a 12-year affiliation with Venrock. Initially he was an investor from 2006 to 2013, when he cofounded and served as a board member of Receptos Pharmaceuticals (acquired by Celgene for more than $7 billion). Dr. Aslan led Venrock's investment in Zeltiq (acquired by Allergan for more than $2 billion) and was involved in the early formation of Fate Therapeutics. Subsequently as an entrepreneur from 2013 to 2018, he was CEO of Adavium Medical, a Brazilian medical device company, which he grew from zero to 350 employees, sales of over US$40 million, and fully integrated R&D, manufacturing, and commercial capabilities. Prior to Venrock, Dr. Aslan was Director of Business Development and Head of Investor Relations for CuraGen, a Nasdaq-listed oncology-focused biotech company. Prior to CuraGen, he was a consultant at Boston Consulting Group (BCG). Dr. Aslan holds a B.S. in biology from Duke University, an M.D. from Yale School of Medicine, and an MBA from Harvard Business School.

Episode SummaryFrontline's Brennan O'Donnell has spent two decades helping companies expand across borders, first as an operator at Google and later as a growth investor backing Series B to D businesses. In this episode, Cheryl and Maxine unpack what's shifted at growth stage in the last 12 months, why the market is still a barbell of “hot or not” deals, and how AI is finally producing application layer companies mature enough for growth rounds.They go deep on Frontline's transatlantic model: seed investing across Europe to help founders raise a Series A and enter the US earlier, and growth investing in the US to help companies expand into Europe with a hands on, concentrated portfolio approach. Brennan breaks down the four pillars Frontline uses to drive international expansion timing, go to market, talent and org design, and location plus the biggest traps founders fall into, like trying to launch in too many markets at once or optimizing for revenue targets instead of learning.You'll also hear why the UK and Ireland are the default first step for 97 percent of US companies entering Europe, when Europe becomes a CEO level priority, how relationship driven sales cycles vary across countries, and why developer led community building can beat traditional sales led expansion for certain AI products. Brennan closes with his Big Cojones moment: moving to the Bay Area for a temporary Google job with everything in storage, then doing it again to help build Google's European HQ in Dublin.Time Stamps03:14 Brennan's first investment: Mode Analytics and a lawn mowing business in Texas06:49 What's changed at growth stage and why “growth” is a different world08:30 Why AI enablement came first and app layer is finally ready for Series B plus10:10 The new risk: fast revenue that's concentrated and not yet durable14:22 Frontline's model: Europe seed plus US growth and why it's unique15:58 What Frontline looks for: category leaders and a line of sight to a 5x outcome16:20 The rough revenue range where growth starts paying attention23:22 The four pillars of expansion: timing, go to market, talent, location26:00 Timing: the 10 percent pull, exec maturity, and why waiting too long is risky29:36 Why Europe expansion has to be a CEO level company priority38:04 Build or buy: why most companies compete into new markets rather than acquire39:10 Developer community expansion as a new go to market wedge41:44 Market selection: why nearly everyone starts with London or Dublin43:56 “Success amnesia” and why you must optimize for learning not quotas48:28 Relationship driven sales cycles and how Europe varies market to market52:43 Big Cojones moment: taking a temp Google job and betting on himself54:26 Doing it again: moving to Dublin in three weeks to help build Google EuropeFirst Cheque is part of Day One.Day One helps founders and startup operators make better business decisions more often. To learn more, join our newsletter to be notified of new First Cheque episodes and upcoming shows.This podcast uses the following third-party services for analysis: Podtrac - https://analytics.podtrac.com/privacy-policy-gdrpSpotify Ad Analytics - https://www.spotify.com/us/legal/ad-analytics-privacy-policy/

From Palantir and Two Sigma to building Goodfire into the poster-child for actionable mechanistic interpretability, Mark Bissell (Member of Technical Staff) and Myra Deng (Head of Product) are trying to turn “peeking inside the model” into a repeatable production workflow by shipping APIs, landing real enterprise deployments, and now scaling the bet with a recent $150M Series B funding round at a $1.25B valuation.In this episode, we go far beyond the usual “SAEs are cool” take. We talk about Goodfire's core bet: that the AI lifecycle is still fundamentally broken because the only reliable control we have is data and we post-train, RLHF, and fine-tune by “slurping supervision through a straw,” hoping the model picks up the right behaviors while quietly absorbing the wrong ones. Goodfire's answer is to build a bi-directional interface between humans and models: read what's happening inside, edit it surgically, and eventually use interpretability during training so customization isn't just brute-force guesswork.Mark and Myra walk through what that looks like when you stop treating interpretability like a lab demo and start treating it like infrastructure: lightweight probes that add near-zero latency, token-level safety filters that can run at inference time, and interpretability workflows that survive messy constraints (multilingual inputs, synthetic→real transfer, regulated domains, no access to sensitive data). We also get a live window into what “frontier-scale interp” means operationally (i.e. steering a trillion-parameter model in real time by targeting internal features) plus why the same tooling generalizes cleanly from language models to genomics, medical imaging, and “pixel-space” world models.We discuss:* Myra + Mark's path: Palantir (health systems, forward-deployed engineering) → Goodfire early team; Two Sigma → Head of Product, translating frontier interpretability research into a platform and real-world deployments* What “interpretability” actually means in practice: not just post-hoc poking, but a broader “science of deep learning” approach across the full AI lifecycle (data curation → post-training → internal representations → model design)* Why post-training is the first big wedge: “surgical edits” for unintended behaviors likereward hacking, sycophancy, noise learned during customization plus the dream of targeted unlearning and bias removal without wrecking capabilities* SAEs vs probes in the real world: why SAE feature spaces sometimes underperform classifiers trained on raw activations for downstream detection tasks (hallucination, harmful intent, PII), and what that implies about “clean concept spaces”* Rakuten in production: deploying interpretability-based token-level PII detection at inference time to prevent routing private data to downstream providers plus the gnarly constraints: no training on real customer PII, synthetic→real transfer, English + Japanese, and tokenization quirks* Why interp can be operationally cheaper than LLM-judge guardrails: probes are lightweight, low-latency, and don't require hosting a second large model in the loop* Real-time steering at frontier scale: a demo of steering Kimi K2 (~1T params) live and finding features via SAE pipelines, auto-labeling via LLMs, and toggling a “Gen-Z slang” feature across multiple layers without breaking tool use* Hallucinations as an internal signal: the case that models have latent uncertainty / “user-pleasing” circuitry you can detect and potentially mitigate more directly than black-box methods* Steering vs prompting: the emerging view that activation steering and in-context learning are more closely connected than people think, including work mapping between the two (even for jailbreak-style behaviors)* Interpretability for science: using the same tooling across domains (genomics, medical imaging, materials) to debug spurious correlations and extract new knowledge up to and including early biomarker discovery work with major partners* World models + “pixel-space” interpretability: why vision/video models make concepts easier to see, how that accelerates the feedback loop, and why robotics/world-model partners are especially interesting design partners* The north star: moving from “data in, weights out” to intentional model design where experts can impart goals and constraints directly, not just via reward signals and brute-force post-training—Goodfire AI* Website: https://goodfire.ai* LinkedIn: https://www.linkedin.com/company/goodfire-ai/* X: https://x.com/GoodfireAIMyra Deng* Website: https://myradeng.com/* LinkedIn: https://www.linkedin.com/in/myra-deng/* X: https://x.com/myra_dengMark Bissell* LinkedIn: https://www.linkedin.com/in/mark-bissell/* X: https://x.com/MarkMBissellFull Video EpisodeTimestamps00:00:00 Introduction00:00:05 Introduction to the Latent Space Podcast and Guests from Goodfire00:00:29 What is Goodfire? Mission and Focus on Interpretability00:01:01 Goodfire's Practical Approach to Interpretability00:01:37 Goodfire's Series B Fundraise Announcement00:02:04 Backgrounds of Mark and Myra from Goodfire00:02:51 Team Structure and Roles at Goodfire00:05:13 What is Interpretability? Definitions and Techniques00:05:30 Understanding Errors00:07:29 Post-training vs. Pre-training Interpretability Applications00:08:51 Using Interpretability to Remove Unwanted Behaviors00:10:09 Grokking, Double Descent, and Generalization in Models00:10:15 404 Not Found Explained00:12:06 Subliminal Learning and Hidden Biases in Models00:14:07 How Goodfire Chooses Research Directions and Projects00:15:00 Troubleshooting Errors00:16:04 Limitations of SAEs and Probes in Interpretability00:18:14 Rakuten Case Study: Production Deployment of Interpretability00:20:45 Conclusion00:21:12 Efficiency Benefits of Interpretability Techniques00:21:26 Live Demo: Real-Time Steering in a Trillion Parameter Model00:25:15 How Steering Features are Identified and Labeled00:26:51 Detecting and Mitigating Hallucinations Using Interpretability00:31:20 Equivalence of Activation Steering and Prompting00:34:06 Comparing Steering with Fine-Tuning and LoRA Techniques00:36:04 Model Design and the Future of Intentional AI Development00:38:09 Getting Started in Mechinterp: Resources, Programs, and Open Problems00:40:51 Industry Applications and the Rise of Mechinterp in Practice00:41:39 Interpretability for Code Models and Real-World Usage00:43:07 Making Steering Useful for More Than Stylistic Edits00:46:17 Applying Interpretability to Healthcare and Scientific Discovery00:49:15 Why Interpretability is Crucial in High-Stakes Domains like Healthcare00:52:03 Call for Design Partners Across Domains00:54:18 Interest in World Models and Visual Interpretability00:57:22 Sci-Fi Inspiration: Ted Chiang and Interpretability01:00:14 Interpretability, Safety, and Alignment Perspectives01:04:27 Weak-to-Strong Generalization and Future Alignment Challenges01:05:38 Final Thoughts and Hiring/Collaboration Opportunities at GoodfireTranscriptShawn Wang [00:00:05]: So welcome to the Latent Space pod. We're back in the studio with our special MechInterp co-host, Vibhu. Welcome. Mochi, Mochi's special co-host. And Mochi, the mechanistic interpretability doggo. We have with us Mark and Myra from Goodfire. Welcome. Thanks for having us on. Maybe we can sort of introduce Goodfire and then introduce you guys. How do you introduce Goodfire today?Myra Deng [00:00:29]: Yeah, it's a great question. So Goodfire, we like to say, is an AI research lab that focuses on using interpretability to understand, learn from, and design AI models. And we really believe that interpretability will unlock the new generation, next frontier of safe and powerful AI models. That's our description right now, and I'm excited to dive more into the work we're doing to make that happen.Shawn Wang [00:00:55]: Yeah. And there's always like the official description. Is there an understatement? Is there an unofficial one that sort of resonates more with a different audience?Mark Bissell [00:01:01]: Well, being an AI research lab that's focused on interpretability, there's obviously a lot of people have a lot that they think about when they think of interpretability. And I think we have a pretty broad definition of what that means and the types of places that can be applied. And in particular, applying it in production scenarios, in high stakes industries, and really taking it sort of from the research world into the real world. Which, you know. It's a new field, so that hasn't been done all that much. And we're excited about actually seeing that sort of put into practice.Shawn Wang [00:01:37]: Yeah, I would say it wasn't too long ago that Anthopic was like still putting out like toy models or superposition and that kind of stuff. And I wouldn't have pegged it to be this far along. When you and I talked at NeurIPS, you were talking a little bit about your production use cases and your customers. And then not to bury the lead, today we're also announcing the fundraise, your Series B. $150 million. $150 million at a 1.25B valuation. Congrats, Unicorn.Mark Bissell [00:02:02]: Thank you. Yeah, no, things move fast.Shawn Wang [00:02:04]: We were talking to you in December and already some big updates since then. Let's dive, I guess, into a bit of your backgrounds as well. Mark, you were at Palantir working on health stuff, which is really interesting because the Goodfire has some interesting like health use cases. I don't know how related they are in practice.Mark Bissell [00:02:22]: Yeah, not super related, but I don't know. It was helpful context to know what it's like. Just to work. Just to work with health systems and generally in that domain. Yeah.Shawn Wang [00:02:32]: And Mara, you were at Two Sigma, which actually I was also at Two Sigma back in the day. Wow, nice.Myra Deng [00:02:37]: Did we overlap at all?Shawn Wang [00:02:38]: No, this is when I was briefly a software engineer before I became a sort of developer relations person. And now you're head of product. What are your sort of respective roles, just to introduce people to like what all gets done in Goodfire?Mark Bissell [00:02:51]: Yeah, prior to Goodfire, I was at Palantir for about three years as a forward deployed engineer, now a hot term. Wasn't always that way. And as a technical lead on the health care team and at Goodfire, I'm a member of the technical staff. And honestly, that I think is about as specific as like as as I could describe myself because I've worked on a range of things. And, you know, it's it's a fun time to be at a team that's still reasonably small. I think when I joined one of the first like ten employees, now we're above 40, but still, it looks like there's always a mix of research and engineering and product and all of the above. That needs to get done. And I think everyone across the team is, you know, pretty, pretty switch hitter in the roles they do. So I think you've seen some of the stuff that I worked on related to image models, which was sort of like a research demo. More recently, I've been working on our scientific discovery team with some of our life sciences partners, but then also building out our core platform for more of like flexing some of the kind of MLE and developer skills as well.Shawn Wang [00:03:53]: Very generalist. And you also had like a very like a founding engineer type role.Myra Deng [00:03:58]: Yeah, yeah.Shawn Wang [00:03:59]: So I also started as I still am a member of technical staff, did a wide range of things from the very beginning, including like finding our office space and all of this, which is we both we both visited when you had that open house thing. It was really nice.Myra Deng [00:04:13]: Thank you. Thank you. Yeah. Plug to come visit our office.Shawn Wang [00:04:15]: It looked like it was like 200 people. It has room for 200 people. But you guys are like 10.Myra Deng [00:04:22]: For a while, it was very empty. But yeah, like like Mark, I spend. A lot of my time as as head of product, I think product is a bit of a weird role these days, but a lot of it is thinking about how do we take our frontier research and really apply it to the most important real world problems and how does that then translate into a platform that's repeatable or a product and working across, you know, the engineering and research teams to make that happen and also communicating to the world? Like, what is interpretability? What is it used for? What is it good for? Why is it so important? All of these things are part of my day-to-day as well.Shawn Wang [00:05:01]: I love like what is things because that's a very crisp like starting point for people like coming to a field. They all do a fun thing. Vibhu, why don't you want to try tackling what is interpretability and then they can correct us.Vibhu Sapra [00:05:13]: Okay, great. So I think like one, just to kick off, it's a very interesting role to be head of product, right? Because you guys, at least as a lab, you're more of an applied interp lab, right? Which is pretty different than just normal interp, like a lot of background research. But yeah. You guys actually ship an API to try these things. You have Ember, you have products around it, which not many do. Okay. What is interp? So basically you're trying to have an understanding of what's going on in model, like in the model, in the internal. So different approaches to do that. You can do probing, SAEs, transcoders, all this stuff. But basically you have an, you have a hypothesis. You have something that you want to learn about what's happening in a model internals. And then you're trying to solve that from there. You can do stuff like you can, you know, you can do activation mapping. You can try to do steering. There's a lot of stuff that you can do, but the key question is, you know, from input to output, we want to have a better understanding of what's happening and, you know, how can we, how can we adjust what's happening on the model internals? How'd I do?Mark Bissell [00:06:12]: That was really good. I think that was great. I think it's also a, it's kind of a minefield of a, if you ask 50 people who quote unquote work in interp, like what is interpretability, you'll probably get 50 different answers. And. Yeah. To some extent also like where, where good fire sits in the space. I think that we're an AI research company above all else. And interpretability is a, is a set of methods that we think are really useful and worth kind of specializing in, in order to accomplish the goals we want to accomplish. But I think we also sort of see some of the goals as even more broader as, as almost like the science of deep learning and just taking a not black box approach to kind of any part of the like AI development life cycle, whether that. That means using interp for like data curation while you're training your model or for understanding what happened during post-training or for the, you know, understanding activations and sort of internal representations, what is in there semantically. And then a lot of sort of exciting updates that were, you know, are sort of also part of the, the fundraise around bringing interpretability to training, which I don't think has been done all that much before. A lot of this stuff is sort of post-talk poking at models as opposed to. To actually using this to intentionally design them.Shawn Wang [00:07:29]: Is this post-training or pre-training or is that not a useful.Myra Deng [00:07:33]: Currently focused on post-training, but there's no reason the techniques wouldn't also work in pre-training.Shawn Wang [00:07:38]: Yeah. It seems like it would be more active, applicable post-training because basically I'm thinking like rollouts or like, you know, having different variations of a model that you can tweak with the, with your steering. Yeah.Myra Deng [00:07:50]: And I think in a lot of the news that you've seen in, in, on like Twitter or whatever, you've seen a lot of unintended. Side effects come out of post-training processes, you know, overly sycophantic models or models that exhibit strange reward hacking behavior. I think these are like extreme examples. There's also, you know, very, uh, mundane, more mundane, like enterprise use cases where, you know, they try to customize or post-train a model to do something and it learns some noise or it doesn't appropriately learn the target task. And a big question that we've always had is like, how do you use your understanding of what the model knows and what it's doing to actually guide the learning process?Shawn Wang [00:08:26]: Yeah, I mean, uh, you know, just to anchor this for people, uh, one of the biggest controversies of last year was 4.0 GlazeGate. I've never heard of GlazeGate. I didn't know that was what it was called. The other one, they called it that on the blog post and I was like, well, how did OpenAI call it? Like officially use that term. And I'm like, that's funny, but like, yeah, I guess it's the pitch that if they had worked a good fire, they wouldn't have avoided it. Like, you know what I'm saying?Myra Deng [00:08:51]: I think so. Yeah. Yeah.Mark Bissell [00:08:53]: I think that's certainly one of the use cases. I think. Yeah. Yeah. I think the reason why post-training is a place where this makes a lot of sense is a lot of what we're talking about is surgical edits. You know, you want to be able to have expert feedback, very surgically change how your model is doing, whether that is, you know, removing a certain behavior that it has. So, you know, one of the things that we've been looking at or is, is another like common area where you would want to make a somewhat surgical edit is some of the models that have say political bias. Like you look at Quen or, um, R1 and they have sort of like this CCP bias.Shawn Wang [00:09:27]: Is there a CCP vector?Mark Bissell [00:09:29]: Well, there's, there are certainly internal, yeah. Parts of the representation space where you can sort of see where that lives. Yeah. Um, and you want to kind of, you know, extract that piece out.Shawn Wang [00:09:40]: Well, I always say, you know, whenever you find a vector, a fun exercise is just like, make it very negative to see what the opposite of CCP is.Mark Bissell [00:09:47]: The super America, bald eagles flying everywhere. But yeah. So in general, like lots of post-training tasks where you'd want to be able to, to do that. Whether it's unlearning a certain behavior or, you know, some of the other kind of cases where this comes up is, are you familiar with like the, the grokking behavior? I mean, I know the machine learning term of grokking.Shawn Wang [00:10:09]: Yeah.Mark Bissell [00:10:09]: Sort of this like double descent idea of, of having a model that is able to learn a generalizing, a generalizing solution, as opposed to even if memorization of some task would suffice, you want it to learn the more general way of doing a thing. And so, you know, another. A way that you can think about having surgical access to a model's internals would be learn from this data, but learn in the right way. If there are many possible, you know, ways to, to do that. Can make interp solve the double descent problem?Shawn Wang [00:10:41]: Depends, I guess, on how you. Okay. So I, I, I viewed that double descent as a problem because then you're like, well, if the loss curves level out, then you're done, but maybe you're not done. Right. Right. But like, if you actually can interpret what is a generalizing or what you're doing. What is, what is still changing, even though the loss is not changing, then maybe you, you can actually not view it as a double descent problem. And actually you're just sort of translating the space in which you view loss and like, and then you have a smooth curve. Yeah.Mark Bissell [00:11:11]: I think that's certainly like the domain of, of problems that we're, that we're looking to get.Shawn Wang [00:11:15]: Yeah. To me, like double descent is like the biggest thing to like ML research where like, if you believe in scaling, then you don't need, you need to know where to scale. And. But if you believe in double descent, then you don't, you don't believe in anything where like anything levels off, like.Vibhu Sapra [00:11:30]: I mean, also tendentially there's like, okay, when you talk about the China vector, right. There's the subliminal learning work. It was from the anthropic fellows program where basically you can have hidden biases in a model. And as you distill down or, you know, as you train on distilled data, those biases always show up, even if like you explicitly try to not train on them. So, you know, it's just like another use case of. Okay. If we can interpret what's happening in post-training, you know, can we clear some of this? Can we even determine what's there? Because yeah, it's just like some worrying research that's out there that shows, you know, we really don't know what's going on.Mark Bissell [00:12:06]: That is. Yeah. I think that's the biggest sentiment that we're sort of hoping to tackle. Nobody knows what's going on. Right. Like subliminal learning is just an insane concept when you think about it. Right. Train a model on not even the logits, literally the output text of a bunch of random numbers. And now your model loves owls. And you see behaviors like that, that are just, they defy, they defy intuition. And, and there are mathematical explanations that you can get into, but. I mean.Shawn Wang [00:12:34]: It feels so early days. Objectively, there are a sequence of numbers that are more owl-like than others. There, there should be.Mark Bissell [00:12:40]: According to, according to certain models. Right. It's interesting. I think it only applies to models that were initialized from the same starting Z. Usually, yes.Shawn Wang [00:12:49]: But I mean, I think that's a, that's a cheat code because there's not enough compute. But like if you believe in like platonic representation, like probably it will transfer across different models as well. Oh, you think so?Mark Bissell [00:13:00]: I think of it more as a statistical artifact of models initialized from the same seed sort of. There's something that is like path dependent from that seed that might cause certain overlaps in the latent space and then sort of doing this distillation. Yeah. Like it pushes it towards having certain other tendencies.Vibhu Sapra [00:13:24]: Got it. I think there's like a bunch of these open-ended questions, right? Like you can't train in new stuff during the RL phase, right? RL only reorganizes weights and you can only do stuff that's somewhat there in your base model. You're not learning new stuff. You're just reordering chains and stuff. But okay. My broader question is when you guys work at an interp lab, how do you decide what to work on and what's kind of the thought process? Right. Because we can ramble for hours. Okay. I want to know this. I want to know that. But like, how do you concretely like, you know, what's the workflow? Okay. There's like approaches towards solving a problem, right? I can try prompting. I can look at chain of thought. I can train probes, SAEs. But how do you determine, you know, like, okay, is this going anywhere? Like, do we have set stuff? Just, you know, if you can help me with all that. Yeah.Myra Deng [00:14:07]: It's a really good question. I feel like we've always at the very beginning of the company thought about like, let's go and try to learn what isn't working in machine learning today. Whether that's talking to customers or talking to researchers at other labs, trying to understand both where the frontier is going and where things are really not falling apart today. And then developing a perspective on how we can push the frontier using interpretability methods. And so, you know, even our chief scientist, Tom, spends a lot of time talking to customers and trying to understand what real world problems are and then taking that back and trying to apply the current state of the art to those problems and then seeing where they fall down basically. And then using those failures or those shortcomings to understand what hills to climb when it comes to interpretability research. So like on the fundamental side, for instance, when we have done some work applying SAEs and probes, we've encountered, you know, some shortcomings in SAEs that we found a little bit surprising. And so have gone back to the drawing board and done work on that. And then, you know, we've done some work on better foundational interpreter models. And a lot of our team's research is focused on what is the next evolution beyond SAEs, for instance. And then when it comes to like control and design of models, you know, we tried steering with our first API and realized that it still fell short of black box techniques like prompting or fine tuning. And so went back to the drawing board and we're like, how do we make that not the case and how do we improve it beyond that? And one of our researchers, Ekdeep, who just joined is actually Ekdeep and Atticus are like steering experts and have spent a lot of time trying to figure out like, what is the research that enables us to actually do this in a much more powerful, robust way? So yeah, the answer is like, look at real world problems, try to translate that into a research agenda and then like hill climb on both of those at the same time.Shawn Wang [00:16:04]: Yeah. Mark has the steering CLI demo queued up, which we're going to go into in a sec. But I always want to double click on when you drop hints, like we found some problems with SAEs. Okay. What are they? You know, and then we can go into the demo. Yeah.Myra Deng [00:16:19]: I mean, I'm curious if you have more thoughts here as well, because you've done it in the healthcare domain. But I think like, for instance, when we do things like trying to detect behaviors within models that are harmful or like behaviors that a user might not want to have in their model. So hallucinations, for instance, harmful intent, PII, all of these things. We first tried using SAE probes for a lot of these tasks. So taking the feature activation space from SAEs and then training classifiers on top of that, and then seeing how well we can detect the properties that we might want to detect in model behavior. And we've seen in many cases that probes just trained on raw activations seem to perform better than SAE probes, which is a bit surprising if you think that SAEs are actually also capturing the concepts that you would want to capture cleanly and more surgically. And so that is an interesting observation. I don't think that is like, I'm not down on SAEs at all. I think there are many, many things they're useful for, but we have definitely run into cases where I think the concept space described by SAEs is not as clean and accurate as we would expect it to be for actual like real world downstream performance metrics.Mark Bissell [00:17:34]: Fair enough. Yeah. It's the blessing and the curse of unsupervised methods where you get to peek into the AI's mind. But sometimes you wish that you saw other things when you walked inside there. Although in the PII instance, I think weren't an SAE based approach actually did prove to be the most generalizable?Myra Deng [00:17:53]: It did work well in the case that we published with Rakuten. And I think a lot of the reasons it worked well was because we had a noisier data set. And so actually the blessing of unsupervised learning is that we actually got to get more meaningful, generalizable signal from SAEs when the data was noisy. But in other cases where we've had like good data sets, it hasn't been the case.Shawn Wang [00:18:14]: And just because you named Rakuten and I don't know if we'll get it another chance, like what is the overall, like what is Rakuten's usage or production usage? Yeah.Myra Deng [00:18:25]: So they are using us to essentially guardrail and inference time monitor their language model usage and their agent usage to detect things like PII so that they don't route private user information.Myra Deng [00:18:41]: And so that's, you know, going through all of their user queries every day. And that's something that we deployed with them a few months ago. And now we are actually exploring very early partnerships, not just with Rakuten, but with other people around how we can help with potentially training and customization use cases as well. Yeah.Shawn Wang [00:19:03]: And for those who don't know, like it's Rakuten is like, I think number one or number two e-commerce store in Japan. Yes. Yeah.Mark Bissell [00:19:10]: And I think that use case actually highlights a lot of like what it looks like to deploy things in practice that you don't always think about when you're doing sort of research tasks. So when you think about some of the stuff that came up there that's more complex than your idealized version of a problem, they were encountering things like synthetic to real transfer of methods. So they couldn't train probes, classifiers, things like that on actual customer data of PII. So what they had to do is use synthetic data sets. And then hope that that transfer is out of domain to real data sets. And so we can evaluate performance on the real data sets, but not train on customer PII. So that right off the bat is like a big challenge. You have multilingual requirements. So this needed to work for both English and Japanese text. Japanese text has all sorts of quirks, including tokenization behaviors that caused lots of bugs that caused us to be pulling our hair out. And then also a lot of tasks you'll see. You might make simplifying assumptions if you're sort of treating it as like the easiest version of the problem to just sort of get like general results where maybe you say you're classifying a sentence to say, does this contain PII? But the need that Rakuten had was token level classification so that you could precisely scrub out the PII. So as we learned more about the problem, you're sort of speaking about what that looks like in practice. Yeah. A lot of assumptions end up breaking. And that was just one instance where you. A problem that seems simple right off the bat ends up being more complex as you keep diving into it.Vibhu Sapra [00:20:41]: Excellent. One of the things that's also interesting with Interp is a lot of these methods are very efficient, right? So where you're just looking at a model's internals itself compared to a separate like guardrail, LLM as a judge, a separate model. One, you have to host it. Two, there's like a whole latency. So if you use like a big model, you have a second call. Some of the work around like self detection of hallucination, it's also deployed for efficiency, right? So if you have someone like Rakuten doing it in production live, you know, that's just another thing people should consider.Mark Bissell [00:21:12]: Yeah. And something like a probe is super lightweight. Yeah. It's no extra latency really. Excellent.Shawn Wang [00:21:17]: You have the steering demos lined up. So we were just kind of see what you got. I don't, I don't actually know if this is like the latest, latest or like alpha thing.Mark Bissell [00:21:26]: No, this is a pretty hacky demo from from a presentation that someone else on the team recently gave. So this will give a sense for, for technology. So you can see the steering and action. Honestly, I think the biggest thing that this highlights is that as we've been growing as a company and taking on kind of more and more ambitious versions of interpretability related problems, a lot of that comes to scaling up in various different forms. And so here you're going to see steering on a 1 trillion parameter model. This is Kimi K2. And so it's sort of fun that in addition to the research challenges, there are engineering challenges that we're now tackling. Cause for any of this to be sort of useful in production, you need to be thinking about what it looks like when you're using these methods on frontier models as opposed to sort of like toy kind of model organisms. So yeah, this was thrown together hastily, pretty fragile behind the scenes, but I think it's quite a fun demo. So screen sharing is on. So I've got two terminal sessions pulled up here. On the left is a forked version that we have of the Kimi CLI that we've got running to point at our custom hosted Kimi model. And then on the right is a set up that will allow us to steer on certain concepts. So I should be able to chat with Kimi over here. Tell it hello. This is running locally. So the CLI is running locally, but the Kimi server is running back to the office. Well, hopefully should be, um, that's too much to run on that Mac. Yeah. I think it's, uh, it takes a full, like each 100 node. I think it's like, you can. You can run it on eight GPUs, eight 100. So, so yeah, Kimi's running. We can ask it a prompt. It's got a forked version of our, uh, of the SG line code base that we've been working on. So I'm going to tell it, Hey, this SG line code base is slow. I think there's a bug. Can you try to figure it out? There's a big code base, so it'll, it'll spend some time doing this. And then on the right here, I'm going to initialize in real time. Some steering. Let's see here.Mark Bissell [00:23:33]: searching for any. Bugs. Feature ID 43205.Shawn Wang [00:23:38]: Yeah.Mark Bissell [00:23:38]: 20, 30, 40. So let me, uh, this is basically a feature that we found that inside Kimi seems to cause it to speak in Gen Z slang. And so on the left, it's still sort of thinking normally it might take, I don't know, 15 seconds for this to kick in, but then we're going to start hopefully seeing him do this code base is massive for real. So we're going to start. We're going to start seeing Kimi transition as the steering kicks in from normal Kimi to Gen Z Kimi and both in its chain of thought and its actual outputs.Mark Bissell [00:24:19]: And interestingly, you can see, you know, it's still able to call tools, uh, and stuff. It's um, it's purely sort of it's it's demeanor. And there are other features that we found for interesting things like concision. So that's more of a practical one. You can make it more concise. Um, the types of programs, uh, programming languages that uses, but yeah, as we're seeing it come in. Pretty good. Outputs.Shawn Wang [00:24:43]: Scheduler code is actually wild.Vibhu Sapra [00:24:46]: Yo, this code is actually insane, bro.Vibhu Sapra [00:24:53]: What's the process of training in SAE on this, or, you know, how do you label features? I know you guys put out a pretty cool blog post about, um, finding this like autonomous interp. Um, something. Something about how agents for interp is different than like coding agents. I don't know while this is spewing up, but how, how do we find feature 43, two Oh five. Yeah.Mark Bissell [00:25:15]: So in this case, um, we, our platform that we've been building out for a long time now supports all the sort of classic out of the box interp techniques that you might want to have like SAE training, probing things of that kind, I'd say the techniques for like vanilla SAEs are pretty well established now where. You take your model that you're interpreting, run a whole bunch of data through it, gather activations, and then yeah, pretty straightforward pipeline to train an SAE. There are a lot of different varieties. There's top KSAEs, batch top KSAEs, um, normal ReLU SAEs. And then once you have your sparse features to your point, assigning labels to them to actually understand that this is a gen Z feature, that's actually where a lot of the kind of magic happens. Yeah. And the most basic standard technique is look at all of your d input data set examples that cause this feature to fire most highly. And then you can usually pick out a pattern. So for this feature, If I've run a diverse enough data set through my model feature 43, two Oh five. Probably tends to fire on all the tokens that sounds like gen Z slang. You know, that's the, that's the time of year to be like, Oh, I'm in this, I'm in this Um, and, um, so, you know, you could have a human go through all 43,000 concepts andVibhu Sapra [00:26:34]: And I've got to ask the basic question, you know, can we get examples where it hallucinates, pass it through, see what feature activates for hallucinations? Can I just, you know, turn hallucination down?Myra Deng [00:26:51]: Oh, wow. You really predicted a project we're already working on right now, which is detecting hallucinations using interpretability techniques. And this is interesting because hallucinations is something that's very hard to detect. And it's like a kind of a hairy problem and something that black box methods really struggle with. Whereas like Gen Z, you could always train a simple classifier to detect that hallucinations is harder. But we've seen that models internally have some... Awareness of like uncertainty or some sort of like user pleasing behavior that leads to hallucinatory behavior. And so, yeah, we have a project that's trying to detect that accurately. And then also working on mitigating the hallucinatory behavior in the model itself as well.Shawn Wang [00:27:39]: Yeah, I would say most people are still at the level of like, oh, I would just turn temperature to zero and that turns off hallucination. And I'm like, well, that's a fundamental misunderstanding of how this works. Yeah.Mark Bissell [00:27:51]: Although, so part of what I like about that question is you, there are SAE based approaches that might like help you get at that. But oftentimes the beauty of SAEs and like we said, the curse is that they're unsupervised. So when you have a behavior that you deliberately would like to remove, and that's more of like a supervised task, often it is better to use something like probes and specifically target the thing that you're interested in reducing as opposed to sort of like hoping that when you fragment the latent space, one of the vectors that pops out.Vibhu Sapra [00:28:20]: And as much as we're training an autoencoder to be sparse, we're not like for sure certain that, you know, we will get something that just correlates to hallucination. You'll probably split that up into 20 other things and who knows what they'll be.Mark Bissell [00:28:36]: Of course. Right. Yeah. So there's no sort of problems with like feature splitting and feature absorption. And then there's the off target effects, right? Ideally, you would want to be very precise where if you reduce the hallucination feature, suddenly maybe your model can't write. Creatively anymore. And maybe you don't like that, but you want to still stop it from hallucinating facts and figures.Shawn Wang [00:28:55]: Good. So Vibhu has a paper to recommend there that we'll put in the show notes. But yeah, I mean, I guess just because your demo is done, any any other things that you want to highlight or any other interesting features you want to show?Mark Bissell [00:29:07]: I don't think so. Yeah. Like I said, this is a pretty small snippet. I think the main sort of point here that I think is exciting is that there's not a whole lot of inter being applied to models quite at this scale. You know, Anthropic certainly has some some. Research and yeah, other other teams as well. But it's it's nice to see these techniques, you know, being put into practice. I think not that long ago, the idea of real time steering of a trillion parameter model would have sounded.Shawn Wang [00:29:33]: Yeah. The fact that it's real time, like you started the thing and then you edited the steering vector.Vibhu Sapra [00:29:38]: I think it's it's an interesting one TBD of what the actual like production use case would be on that, like the real time editing. It's like that's the fun part of the demo, right? You can kind of see how this could be served behind an API, right? Like, yes, you're you only have so many knobs and you can just tweak it a bit more. And I don't know how it plays in. Like people haven't done that much with like, how does this work with or without prompting? Right. How does this work with fine tuning? Like, there's a whole hype of continual learning, right? So there's just so much to see. Like, is this another parameter? Like, is it like parameter? We just kind of leave it as a default. We don't use it. So I don't know. Maybe someone here wants to put out a guide on like how to use this with prompting when to do what?Mark Bissell [00:30:18]: Oh, well, I have a paper recommendation. I think you would love from Act Deep on our team, who is an amazing researcher, just can't say enough amazing things about Act Deep. But he actually has a paper that as well as some others from the team and elsewhere that go into the essentially equivalence of activation steering and in context learning and how those are from a he thinks of everything in a cognitive neuroscience Bayesian framework, but basically how you can precisely show how. Prompting in context, learning and steering exhibit similar behaviors and even like get quantitative about the like magnitude of steering you would need to do to induce a certain amount of behavior similar to certain prompting, even for things like jailbreaks and stuff. It's a really cool paper. Are you saying steering is less powerful than prompting? More like you can almost write a formula that tells you how to convert between the two of them.Myra Deng [00:31:20]: And so like formally equivalent actually in the in the limit. Right.Mark Bissell [00:31:24]: So like one case study of this is for jailbreaks there. I don't know. Have you seen the stuff where you can do like many shot jailbreaking? You like flood the context with examples of the behavior. And the topic put out that paper.Shawn Wang [00:31:38]: A lot of people were like, yeah, we've been doing this, guys.Mark Bissell [00:31:40]: Like, yeah, what's in this in context learning and activation steering equivalence paper is you can like predict the number. Number of examples that you will need to put in there in order to jailbreak the model. That's cool. By doing steering experiments and using this sort of like equivalence mapping. That's cool. That's really cool. It's very neat. Yeah.Shawn Wang [00:32:02]: I was going to say, like, you know, I can like back rationalize that this makes sense because, you know, what context is, is basically just, you know, it updates the KV cache kind of and like and then every next token inference is still like, you know, the sheer sum of everything all the way. It's plus all the context. It's up to date. And you could, I guess, theoretically steer that with you probably replace that with your steering. The only problem is steering typically is on one layer, maybe three layers like like you did. So it's like not exactly equivalent.Mark Bissell [00:32:33]: Right, right. There's sort of you need to get precise about, yeah, like how you sort of define steering and like what how you're modeling the setup. But yeah, I've got the paper pulled up here. Belief dynamics reveal the dual nature. Yeah. The title is Belief Dynamics Reveal the Dual Nature of Incompetence. And it's an exhibition of the practical context learning and activation steering. So Eric Bigelow, Dan Urgraft on the who are doing fellowships at Goodfire, Ekt Deep's the final author there.Myra Deng [00:32:59]: I think actually to your question of like, what is the production use case of steering? I think maybe if you just think like one level beyond steering as it is today. Like imagine if you could adapt your model to be, you know, an expert legal reasoner. Like in almost real time, like very quickly. efficiently using human feedback or using like your semantic understanding of what the model knows and where it knows that behavior. I think that while it's not clear what the product is at the end of the day, it's clearly very valuable. Thinking about like what's the next interface for model customization and adaptation is a really interesting problem for us. Like we have heard a lot of people actually interested in fine-tuning an RL for open weight models in production. And so people are using things like Tinker or kind of like open source libraries to do that, but it's still very difficult to get models fine-tuned and RL'd for exactly what you want them to do unless you're an expert at model training. And so that's like something we'reShawn Wang [00:34:06]: looking into. Yeah. I never thought so. Tinker from Thinking Machines famously uses rank one LoRa. Is that basically the same as steering? Like, you know, what's the comparison there?Mark Bissell [00:34:19]: Well, so in that case, you are still applying updates to the parameters, right?Shawn Wang [00:34:25]: Yeah. You're not touching a base model. You're touching an adapter. It's kind of, yeah.Mark Bissell [00:34:30]: Right. But I guess it still is like more in parameter space then. I guess it's maybe like, are you modifying the pipes or are you modifying the water flowing through the pipes to get what you're after? Yeah. Just maybe one way.Mark Bissell [00:34:44]: I like that analogy. That's my mental map of it at least, but it gets at this idea of model design and intentional design, which is something that we're, that we're very focused on. And just the fact that like, I hope that we look back at how we're currently training models and post-training models and just think what a primitive way of doing that right now. Like there's no intentionalityShawn Wang [00:35:06]: really in... It's just data, right? The only thing in control is what data we feed in.Mark Bissell [00:35:11]: So, so Dan from Goodfire likes to use this analogy of, you know, he has a couple of young kids and he talks about like, what if I could only teach my kids how to be good people by giving them cookies or like, you know, giving them a slap on the wrist if they do something wrong, like not telling them why it was wrong or like what they should have done differently or something like that. Just figure it out. Right. Exactly. So that's RL. Yeah. Right. And, and, you know, it's sample inefficient. There's, you know, what do they say? It's like slurping feedback. It's like, slurping supervision. Right. And so you'd like to get to the point where you can have experts giving feedback to their models that are, uh, internalized and, and, you know, steering is an inference time way of sort of getting that idea. But ideally you're moving to a world whereVibhu Sapra [00:36:04]: it is much more intentional design in perpetuity for these models. Okay. This is one of the questions we asked Emmanuel from Anthropic on the podcast a few months ago. Basically the question, was you're at a research lab that does model training, foundation models, and you're on an interp team. How does it tie back? Right? Like, does this, do ideas come from the pre-training team? Do they go back? Um, you know, so for those interested, you can, you can watch that. There wasn't too much of a connect there, but it's still something, you know, it's something they want toMark Bissell [00:36:33]: push for down the line. It can be useful for all of the above. Like there are certainly post-hocVibhu Sapra [00:36:39]: use cases where it doesn't need to touch that. I think the other thing a lot of people forget is this stuff isn't too computationally expensive, right? Like I would say, if you're interested in getting into research, MechInterp is one of the most approachable fields, right? A lot of this train an essay, train a probe, this stuff, like the budget for this one, there's already a lot done. There's a lot of open source work. You guys have done some too. Um, you know,Shawn Wang [00:37:04]: There's like notebooks from the Gemini team for Neil Nanda or like, this is how you do it. Just step through the notebook.Vibhu Sapra [00:37:09]: Even if you're like, not even technical with any of this, you can still make like progress. There, you can look at different activations, but, uh, if you do want to get into training, you know, training this stuff, correct me if I'm wrong is like in the thousands of dollars, not even like, it's not that high scale. And then same with like, you know, applying it, doing it for post-training or all this stuff is fairly cheap in scale of, okay. I want to get into like model training. I don't have compute for like, you know, pre-training stuff. So it's, it's a very nice field to get into. And also there's a lot of like open questions, right? Um, some of them have to go with, okay, I want a product. I want to solve this. Like there's also just a lot of open-ended stuff that people could work on. That's interesting. Right. I don't know if you guys have any calls for like, what's open questions, what's open work that you either open collaboration with, or like, you'd just like to see solved or just, you know, for people listening that want to get into McInturk because people always talk about it. What are, what are the things they should check out? Start, of course, you know, join you guys as well. I'm sure you're hiring.Myra Deng [00:38:09]: There's a paper, I think from, was it Lee, uh, Sharky? It's open problems and, uh, it's, it's a bit of interpretability, which I recommend everyone who's interested in the field. Read. I'm just like a really comprehensive overview of what are the things that experts in the field think are the most important problems to be solved. I also think to your point, it's been really, really inspiring to see, I think a lot of young people getting interested in interpretability, actually not just young people also like scientists to have been, you know, experts in physics for many years and in biology or things like this, um, transitioning into interp, because the barrier of, of what's now interp. So it's really cool to see a number to entry is, you know, in some ways low and there's a lot of information out there and ways to get started. There's this anecdote of like professors at universities saying that all of a sudden every incoming PhD student wants to study interpretability, which was not the case a few years ago. So it just goes to show how, I guess, like exciting the field is, how fast it's moving, how quick it is to get started and things like that.Mark Bissell [00:39:10]: And also just a very welcoming community. You know, there's an open source McInturk Slack channel. There are people are always posting questions and just folks in the space are always responsive if you ask things on various forums and stuff. But yeah, the open paper, open problems paper is a really good one.Myra Deng [00:39:28]: For other people who want to get started, I think, you know, MATS is a great program. What's the acronym for? Machine Learning and Alignment Theory Scholars? It's like the...Vibhu Sapra [00:39:40]: Normally summer internship style.Myra Deng [00:39:42]: Yeah, but they've been doing it year round now. And actually a lot of our full-time staff have come through that program or gone through that program. And it's great for anyone who is transitioning into interpretability. There's a couple other fellows programs. We do one as well as Anthropic. And so those are great places to get started if anyone is interested.Mark Bissell [00:40:03]: Also, I think been seen as a research field for a very long time. But I think engineering... I think engineers are sorely wanted for interpretability as well, especially at Goodfire, but elsewhere, as it does scale up.Shawn Wang [00:40:18]: I should mention that Lee actually works with you guys, right? And in the London office and I'm adding our first ever McInturk track at AI Europe because I see this industry applications now emerging. And I'm pretty excited to, you know, help push that along. Yeah, I was looking forward to that. It'll effectively be the first industry McInturk conference. Yeah. I'm so glad you added that. You know, it's still a little bit of a bet. It's not that widespread, but I can definitely see this is the time to really get into it. We want to be early on things.Mark Bissell [00:40:51]: For sure. And I think the field understands this, right? So at ICML, I think the title of the McInturk workshop this year was actionable interpretability. And there was a lot of discussion around bringing it to various domains. Everyone's adding pragmatic, actionable, whatever.Shawn Wang [00:41:10]: It's like, okay, well, we weren't actionable before, I guess. I don't know.Vibhu Sapra [00:41:13]: And I mean, like, just, you know, being in Europe, you see the Interp room. One, like old school conferences, like, I think they had a very tiny room till they got lucky and they got it doubled. But there's definitely a lot of interest, a lot of niche research. So you see a lot of research coming out of universities, students. We covered the paper last week. It's like two unknown authors, not many citations. But, you know, you can make a lot of meaningful work there. Yeah. Yeah. Yeah.Shawn Wang [00:41:39]: Yeah. I think people haven't really mentioned this yet. It's just Interp for code. I think it's like an abnormally important field. We haven't mentioned this yet. The conspiracy theory last two years ago was when the first SAE work came out of Anthropic was they would do like, oh, we just used SAEs to turn the bad code vector down and then turn up the good code. And I think like, isn't that the dream? Like, you know, like, but basically, I guess maybe, why is it funny? Like, it's... If it was realistic, it would not be funny. It would be like, no, actually, we should do this. But it's funny because we know there's like, we feel there's some limitations to what steering can do. And I think a lot of the public image of steering is like the Gen Z stuff. Like, oh, you can make it really love the Golden Gate Bridge, or you can make it speak like Gen Z. To like be a legal reasoner seems like a huge stretch. Yeah. And I don't know if that will get there this way. Yeah.Myra Deng [00:42:36]: I think, um, I will say we are announcing. Something very soon that I will not speak too much about. Um, but I think, yeah, this is like what we've run into again and again is like, we, we don't want to be in the world where steering is only useful for like stylistic things. That's definitely not, not what we're aiming for. But I think the types of interventions that you need to do to get to things like legal reasoning, um, are much more sophisticated and require breakthroughs in, in learning algorithms. And that's, um...Shawn Wang [00:43:07]: And is this an emergent property of scale as well?Myra Deng [00:43:10]: I think so. Yeah. I mean, I think scale definitely helps. I think scale allows you to learn a lot of information and, and reduce noise across, you know, large amounts of data. But I also think we think that there's ways to do things much more effectively, um, even, even at scale. So like actually learning exactly what you want from the data and not learning things that you do that you don't want exhibited in the data. So we're not like anti-scale, but we are also realizing that scale is not going to get us anywhere. It's not going to get us to the type of AI development that we want to be at in, in the future as these models get more powerful and get deployed in all these sorts of like mission critical contexts. Current life cycle of training and deploying and evaluations is, is to us like deeply broken and has opportunities to, to improve. So, um, more to come on that very, very soon.Mark Bissell [00:44:02]: And I think that that's a use basically, or maybe just like a proof point that these concepts do exist. Like if you can manipulate them in the precise best way, you can get the ideal combination of them that you desire. And steering is maybe the most coarse grained sort of peek at what that looks like. But I think it's evocative of what you could do if you had total surgical control over every concept, every parameter. Yeah, exactly.Myra Deng [00:44:30]: There were like bad code features. I've got it pulled up.Vibhu Sapra [00:44:33]: Yeah. Just coincidentally, as you guys are talking.Shawn Wang [00:44:35]: This is like, this is exactly.Vibhu Sapra [00:44:38]: There's like specifically a code error feature that activates and they show, you know, it's not, it's not typo detection. It's like, it's, it's typos in code. It's not typical typos. And, you know, you can, you can see it clearly activates where there's something wrong in code. And they have like malicious code, code error. They have a whole bunch of sub, you know, sub broken down little grain features. Yeah.Shawn Wang [00:45:02]: Yeah. So, so the, the rough intuition for me, the, why I talked about post-training was that, well, you just, you know, have a few different rollouts with all these things turned off and on and whatever. And then, you know, you can, that's, that's synthetic data you can kind of post-train on. Yeah.Vibhu Sapra [00:45:13]: And I think we make it sound easier than it is just saying, you know, they do the real hard work.Myra Deng [00:45:19]: I mean, you guys, you guys have the right idea. Exactly. Yeah. We replicated a lot of these features in, in our Lama models as well. I remember there was like.Vibhu Sapra [00:45:26]: And I think a lot of this stuff is open, right? Like, yeah, you guys opened yours. DeepMind has opened a lot of essays on Gemma. Even Anthropic has opened a lot of this. There's, there's a lot of resources that, you know, we can probably share of people that want to get involved.Shawn Wang [00:45:41]: Yeah. And special shout out to like Neuronpedia as well. Yes. Like, yeah, amazing piece of work to visualize those things.Myra Deng [00:45:49]: Yeah, exactly.Shawn Wang [00:45:50]: I guess I wanted to pivot a little bit on, onto the healthcare side, because I think that's a big use case for you guys. We haven't really talked about it yet. This is a bit of a crossover for me because we are, we are, we do have a separate science pod that we're starting up for AI, for AI for science, just because like, it's such a huge investment category and also I'm like less qualified to do it, but we actually have bio PhDs to cover that, which is great, but I need to just kind of recover, recap your work, maybe on the evil two stuff, but then, and then building forward.Mark Bissell [00:46:17]: Yeah, for sure. And maybe to frame up the conversation, I think another kind of interesting just lens on interpretability in general is a lot of the techniques that were described. are ways to solve the AI human interface problem. And it's sort of like bidirectional communication is the goal there. So what we've been talking about with intentional design of models and, you know, steering, but also more advanced techniques is having humans impart our desires and control into models and over models. And the reverse is also very interesting, especially as you get to superhuman models, whether that's narrow superintelligence, like these scientific models that work on genomics, data, medical imaging, things like that. But down the line, you know, superintelligence of other forms as well. What knowledge can the AIs teach us as sort of that, that the other direction in that? And so some of our life science work to date has been getting at exactly that question, which is, well, some of it does look like debugging these various life sciences models, understanding if they're actually performing well, on tasks, or if they're picking up on spurious correlations, for instance, genomics models, you would like to know whether they are sort of focusing on the biologically relevant things that you care about, or if it's using some simpler correlate, like the ancestry of the person that it's looking at. But then also in the instances where they are superhuman, and maybe they are understanding elements of the human genome that we don't have names for or specific, you know, yeah, discoveries that they've made that that we don't know about, that's, that's a big goal. And so we're already seeing that, right, we are partnered with organizations like Mayo Clinic, leading research health system in the United States, our Institute, as well as a startup called Prima Menta, which focuses on neurodegenerative disease. And in our partnership with them, we've used foundation models, they've been training and applied our interpretability techniques to find novel biomarkers for Alzheimer's disease. So I think this is just the tip of the iceberg. But it's, that's like a flavor of some of the things that we're working on.Shawn Wang [00:48:36]: Yeah, I think that's really fantastic. Obviously, we did the Chad Zuckerberg pod last year as well. And like, there's a plethora of these models coming out, because there's so much potential and research. And it's like, very interesting how it's basically the same as language models, but just with a different underlying data set. But it's like, it's the same exact techniques. Like, there's no change, basically.Mark Bissell [00:48:59]: Yeah. Well, and even in like other domains, right? Like, you know, robotics, I know, like a lot of the companies just use Gemma as like the like backbone, and then they like make it into a VLA that like takes these actions. It's, it's, it's transformers all the way down. So yeah.Vibhu Sapra [00:49:15]: Like we have Med Gemma now, right? Like this week, even there was Med Gemma 1.5. And they're training it on this stuff, like 3d scans, medical domain knowledge, and all that stuff, too. So there's a push from both sides. But I think the thing that, you know, one of the things about McInturpp is like, you're a little bit more cautious in some domains, right? So healthcare, mainly being one, like guardrails, understanding, you know, we're more risk adverse to something going wrong there. So even just from a basic understanding, like, if we're trusting these systems to make claims, we want to know why and what's going on.Myra Deng [00:49:51]: Yeah, I think there's totally a kind of like deployment bottleneck to actually using. foundation models for real patient usage or things like that. Like, say you're using a model for rare disease prediction, you probably want some explanation as to why your model predicted a certain outcome, and an interpretable explanation at that. So that's definitely a use case. But I also think like, being able to extract scientific information that no human knows to accelerate drug discovery and disease treatment and things like that actually is a really, really big unlock for science, like scientific discovery. And you've seen a lot of startups, like say that they're going to accelerate scientific discovery. And I feel like we actually are doing that through our interp techniques. And kind of like, almost by accident, like, I think we got reached out to very, very early on from these healthcare institutions. And none of us had healthcare.Shawn Wang [00:50:49]: How did they even hear of you? A podcast.Myra Deng [00:50:51]: Oh, okay. Yeah, podcast.Vibhu Sapra [00:50:53]: Okay, well, now's that time, you know.Myra Deng [00:50:55]: Everyone can call us.Shawn Wang [00:50:56]: Podcasts are the most important thing. Everyone should listen to podcasts.Myra Deng [00:50:59]: Yeah, they reached out. They were like, you know, we have these really smart models that we've trained, and we want to know what they're doing. And we were like, really early that time, like three months old, and it was a few of us. And we were like, oh, my God, we've never used these models. Let's figure it out. But it's also like, great proof that interp techniques scale pretty well across domains. We didn't really have to learn too much about.Shawn Wang [00:51:21]: Interp is a machine learning technique, machine learning skills everywhere, right? Yeah. And it's obviously, it's just like a general insight. Yeah. Probably to finance too, I think, which would be fun for our history. I don't know if you have anything to say there.Mark Bissell [00:51:34]: Yeah, well, just across the science. Like, we've also done work on material science. Yeah, it really runs the gamut.Vibhu Sapra [00:51:40]: Yeah. Awesome. And, you know, for those that should reach out, like, you're obviously experts in this, but like, is there a call out for people that you're looking to partner with, design partners, people to use your stuff outside of just, you know, the general developer that wants to. Plug and play steering stuff, like on the research side more so, like, are there ideal design partners, customers, stuff like that?Myra Deng [00:52:03]: Yeah, I can talk about maybe non-life sciences, and then I'm curious to hear from you on the life sciences side. But we're looking for design partners across many domains, language, anyone who's customizing language models or trying to push the frontier of code or reasoning models is really interesting to us. And then also interested in the frontier of modeling. There's a lot of models that work in, like, pixel space, as we call it. So if you're doing world models, video models, even robotics, where there's not a very clean natural language interface to interact with, I think we think that Interp can really help and are looking for a few partners in that space.Shawn Wang [00:52:43]: Just because you mentioned the keyword

Bassem took Briq from a failed data idea to a Series B leader in construction financial automation.But the path wasn't linear. In this episode, Bassem reveals how he pivoted to RPA bots, why he killed a high-growth fintech product to survive the 2023 cash crunch, and how he uses a relentless "Go-to-Market" strategy. He breaks down his exact ABM playbook, why he hates trade shows, and why he believes AI orchestration is a bigger shift than the cloud.Why You Should ListenHow to identify the "Challenger" who will kill your deal.Why trade shows are a waste of money (and what to do instead).The "1-Person Webinar" hack to close high-value accounts.The brutal reality of cutting 50% of staff to survive.Why selling "risk reduction" beats selling "time saved."Keywordsstartup podcast, startup podcast for founders, product market fit, account based marketing, construction tech, go to market strategy, enterprise sales, finding pmf, robotic process automation, ai orchestration00:00:00 Intro00:06:23 The RPA "Aha" Moment with a Tech Giant00:11:52 Selling Risk vs. Selling Time Saved00:13:27 The "New CFO" Signal in Account Based Marketing00:17:06 Identifying the "Challenger" in Enterprise Sales00:23:22 The 1-Person Webinar Strategy00:29:19 Killing the Fintech Product to Survive 202300:36:20 Why You Never Truly Have Product Market FitSend me a message to let me know what you think!

Sponsored by Chargebee, subscription and revenue management → check out their startup offer: https://www.chargebee.com/startupsJake Stauch, Founder of Servalhttps://www.linkedin.com/in/jakestauch/

The investment comes from backers including the Qatar Investment Authority as demand for chips beyond Nvidia soars and as Qatar aims to build out its AI infrastructure. Also, To build and sell more of its autonomous underwater vehicles (AUVs), Apeiron Labs recently closed a $9.5 million Series A round led by Dyne Ventures, RA Capital Management Planetary Health and S2G Investments, the company exclusively told TechCrunch. Assembly Ventures, Bay Bridge Ventures, and TFX Capital participated. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Host Isaiah Buchanan kicks off this Tuesday edition with a significant legal victory for NFI CEO Sidney Brown. An appellate court has affirmed the dismissal of criminal charges against the executive regarding real estate development rights in Camden, New Jersey. Next, the show examines how robust consumer spending is leading retailers to adopt leaner inventory strategies. While this shift is softening ocean shipping demand, it is expected to drive up domestic truckload rates and tender rejections in the near term. In FreightTech news, startup GenLogs has secured $60 million in Series B funding to expand its AI-powered supply chain intelligence platform. The company aims to use its nationwide sensor network to enhance visibility and combat freight fraud across the industry. Follow the FreightWaves NOW Podcast Other FreightWaves Shows Learn more about your ad choices. Visit megaphone.fm/adchoices

In this lightning installment, we are joined by Ryan Joyce, CEO & Co-Founder of GenLogs. On February 2, 2026, GenLogs announced that it has closed a $60 million Series B funding round. The investment, led by Battery Ventures and including participation from IVP, Cathay Innovation, 9Yards, and existing backers like Venrock, Steel Atlas, HOF Capital, TitletownTech, and Autotech Ventures, brings the company's total funding to $81 million since its founding in 2023. This latest capital raise aims to expand GenLogs' AI-powered Truck Intelligence™ platform, which leverages a nationwide network of roadside sensors, satellites, and diverse data streams to provide unprecedented visibility into U.S. trucking operations. We talk This program is brought to you by DAT Freight & Analytics. Since 1978, DAT has helped truckers & brokers discover more available loads. Whether you're heading home or looking for your next adventure, DAT is building the most trusted marketplace in freight. New users of DAT can save 10% off for the first 12 months by following the link below. Built on the latest technology, DAT One gives you control over every aspect of moving freight, so that you can run your business with speed & efficiency. This program is also brought to you by our newest sponsor, GenLogs. GenLogs is setting a new standard of care for freight intelligence. Book your demo for GenLogs today at www.genlogs.io today!

Host Isaiah Buchanan kicks off this Tuesday edition with a significant legal victory for NFI CEO Sidney Brown. An appellate court has affirmed the dismissal of criminal charges against the executive regarding real estate development rights in Camden, New Jersey. Next, the show examines how robust consumer spending is leading retailers to adopt leaner inventory strategies. While this shift is softening ocean shipping demand, it is expected to drive up domestic truckload rates and tender rejections in the near term. In FreightTech news, startup GenLogs has secured $60 million in Series B funding to expand its AI-powered supply chain intelligence platform. The company aims to use its nationwide sensor network to enhance visibility and combat freight fraud across the industry. Follow the FreightWaves NOW Podcast Other FreightWaves Shows Learn more about your ad choices. Visit megaphone.fm/adchoices

Poland says weak security left parts of its power grid exposed. A Russian-linked hacker alliance threatens Denmark with a promised cyber offensive. Fancy Bear moves fast on a new Microsoft Office flaw, hitting Ukrainian and EU targets. Researchers find a sprawling supply chain attack buried in the ClawdBot AI ecosystem. A new report looks at how threats are shaping the work of journalists and security researchers. A stealthy Windows malware campaign blends Pulsar RAT with Stealerv37. A former Google engineer is convicted of stealing AI trade secrets for China. The latest cybersecurity funding and deal news. On our Afternoon Cyber Tea segment, Microsoft's Ann Johnson chats with Dr. Lorrie Cranor from Carnegie Mellon about security design. The AI dinosaur that knew too much.  Remember to leave us a 5-star rating and review in your favorite podcast app. Miss an episode? Sign-up for our daily intelligence roundup, Daily Briefing, and you'll never miss a beat. And be sure to follow CyberWire Daily on LinkedIn. Afternoon Cyber Tea Dr. Lorrie Cranor⁠, Director of the CyLab Security and Privacy Institute at Carnegie Mellon University joins Ann Johnson, Corporate Vice President, Microsoft, on this month's segment of Afternoon Cyber Tea to discuss the critical gap between security design and real-world usability. They explore why security tools often fail users, the ongoing challenges with passwords and password less authentication, and how privacy expectations have evolved in an era of constant data collection. You can listen to Ann and Lorrie's full conversation here, and catch new episodes Afternoon Cyber Tea every other Tuesday on your favorite podcast app. Selected Reading Russian hackers breached Polish power grid thanks to bad security, report says (TechCrunch) Newly Established Russian Hacker Alliance Threatens Denmark (Truesec) Fancy Bear Exploits Microsoft Office Flaw in Ukraine, EU Cyber-Attacks (Infosecurity Magazine) Notepad++ Hijacked by State-Sponsored Hackers (Notepad++) ClawdBot Skills Just Ganked Your Crypto (OpenSource Malware Blog) Under Pressure: Exploring the effect of legal and criminal threats on security researchers and journalists (DataBreaches.Net) Windows Malware Uses Pulsar RAT for Live Chats While Stealing Data (Hackread) U.S. convicts ex-Google engineer for sending AI tech data to China (Bleeping Computer) Upwind secures $250 million in a Series B round. (N2K Pro Business Briefing)  Don't Buy Internet-Connected Toys For Your Kids (Blackout VPN) Share your feedback. What do you think about CyberWire Daily? Please take a few minutes to share your thoughts with us by completing our brief listener survey. Thank you for helping us continue to improve our show. Want to hear your company in the show? N2K CyberWire helps you reach the industry's most influential leaders and operators, while building visibility, authority, and connectivity across the cybersecurity community. Learn more at sponsor.thecyberwire.com. The CyberWire is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

Send us a textShangyup Kim is the CEO of ZEP QUIZ, the metaverse-based gamified K-12 learning platform. Built AI-driven tools that help teachers motivate students at scale and is expanding across Asia and the U.S. As a former ZEPETO leader, focusing on gamification, futuristic change, and education. Formerly built two Series B companies.Jangwoo Bae is the business developer of ZEP QUIZ, a former English teacher and school administrator in Thailand during COVID-19. Recognized the need for educational tools that extend beyond traditional classrooms and the urgent necessity for digital transformation in schools, while preserving the fundamental values of learning.

Friday Gold Falls 8%, Silver Drops Below $85, Precious Metals Lose $7 TrillionTalos Raises $150 M in Series B, Attracts Strategic Institutional InvestorsSpaceX‑Tesla Merger Talks Spotlight Nearly 20,000 Bitcoins In FocusBinance Blames Macro Shock, Not Exchange Failure, for October's $19 Billion Liquidation Cascade

If you spend any time on the water, you've probably heard the joke that the two best days of a boat owner's life are the day they buy the boat… and the day they sell it.For today's guest, that line isn't just a joke. It's a problem statement.Mitch Lee grew up around boats. He loved being on the water, but he also experienced firsthand how loud, smelly, maintenance-heavy, and frustrating boat ownership can be. And once electric vehicles started proving what was possible on land, one idea kept coming back to him: if electrification makes sense in cars on the road, it might actually make even more sense in boats on the water.Mitch is the co-founder and CEO of Arc, an electric boat company building both consumer and commercial vessels. Arc started in the premium wake sports market, selling directly to consumers and using those early boats to develop its electric propulsion technology. That platform is now being deployed in commercial applications, where electrification can improve reliability, lower costs, and support ports that want to electrify.Founded in 2021, Arc has raised over $110 million through its Series B and has scaled from delivering its first customer boat, which took two years to build, to now producing multiple boats per week, while also landing major commercial partnerships at working ports.In our conversation, Mitch walks me through his path that led to Arc, from a childhood fascination with stocks and compounding interest, to building and selling his first company in the personal finance space, to betting big on electric boats and what electrification can unlock on the water.About Powerhouse Innovation and Powerhouse VenturesPowerhouse Ventures backs seed stage startups developing innovative software to advance clean energy, mobility, and industry. If you are thinking about building something in this space, get in touch with our team.Powerhouse Innovation is a best in class consulting firm, powered by the strongest energy innovation network, data and team in our industry. We partner with world's leading corporations, investors, and utilities to source and evaluate disruptive startups shaping the future of energy and industry.To hear more stories of founders building our energy abundant future, hit the “subscribe” button and leave us a review.

The $250 million Series B was led by Bessemer Venture Partners, with participation from Salesforce Ventures and Picture Capital. Also, Outtake makes an agentic cybersecurity platform to help enterprises detect identity fraud. Its angel investors are a who's who. Learn more about your ad choices. Visit podcastchoices.com/adchoices

The crew breaks down Superstate's massive $82M Series B for tokenization, the explosive rise of TradeXYZ's commodities trading hitting $1B+ volume, different tokenization models from "bootleg" to "back office," the ClawdBot AI phenomenon taking over coding, and how agent-based development is revolutionizing crypto software engineering. Welcome to The Chopping Block — where crypto insiders Haseeb Qureshi, Tom Schmidt, Tarun Chitra, and Robert Leshner chop it up about the latest in crypto. This week, Robert drops news about Superstate's massive $82 million Series B raise led by Bain Capital to bring Wall Street on-chain through tokenization. The crew dives deep into the explosive growth of Hip3 markets, particularly TradeXYZ's commodities trading that's hitting over $1 billion in daily volume as precious metals rip to all-time highs. They break down the different tokenization models emerging - from "bootleg" third-party approaches to "back office" settlement tools to issuer-led official tokenization. Then the conversation shifts to the ClawdBot phenomenon taking the internet by storm, exploring how AI agents are revolutionizing coding and what this means for the future of software engineering in crypto. From vibe coding to the complete transformation of how startups will be built, the hosts examine whether we're witnessing a fundamental shift in how technical work gets done. Show highlights

Tim and Tyler talk with Tim Bucher, Founder and CEO of Agtonomy, about what trends he saw at CES and Agtonomy's recent Series B round. — This episode is presented by Yield Energy. Yield for Growers. — Links Agtonomy - https://www.agtonomy.com 

A new AI lab called Flapping Airplanes launched yesterday, and a Sequoia partner has an interesting take on why they stand out. Also, Upwind's $250 million Series B was led by Bessemer Venture Partners, with participation from Salesforce Ventures and Picture Capital. Learn more about your ad choices. Visit podcastchoices.com/adchoices

The crew breaks down Superstate's massive $82M Series B for tokenization, the explosive rise of TradeXYZ's commodities trading hitting $1B+ volume, different tokenization models from "bootleg" to "back office," the ClawdBot AI phenomenon taking over coding, and how agent-based development is revolutionizing crypto software engineering. Welcome to The Chopping Block — where crypto insiders Haseeb Qureshi, Tom Schmidt, Tarun Chitra, and Robert Leshner chop it up about the latest in crypto. This week, Robert drops news about Superstate's massive $82 million Series B raise led by Bain Capital to bring Wall Street on-chain through tokenization. The crew dives deep into the explosive growth of Hip3 markets, particularly TradeXYZ's commodities trading that's hitting over $1 billion in daily volume as precious metals rip to all-time highs. They break down the different tokenization models emerging - from "bootleg" third-party approaches to "back office" settlement tools to issuer-led official tokenization. Then the conversation shifts to the ClawdBot phenomenon taking the internet by storm, exploring how AI agents are revolutionizing coding and what this means for the future of software engineering in crypto. From vibe coding to the complete transformation of how startups will be built, the hosts examine whether we're witnessing a fundamental shift in how technical work gets done. Show highlights

In this episode, we're joined by Roeland Delrue, Co-founder of Aikido Security, the fast-growing application security platform that just became a unicorn in under 4 years.  Roeland shares what it really takes to scale a company at breakneck speed, from going 5× year-over-year, to balancing startups and enterprise in one GTM motion, to raising a Series B with a single goal: becoming “unignorable.” We unpack how Aikido uses product-led growth, brutal revenue focus, and buyer-first sales mentality to win in one of the most competitive markets in SaaS. We spoke with Roeland about building for continuous wins, why revenue clarity beats buzzwords, and how Aikido joins the buyer's journey instead of forcing rigid sales methodologies. Here are some of the key questions we address: How did Aikido grow from $5M to $20M+ ARR in one year? What does it mean to build an unignorable company? Why brutal focus on revenue simplifies product, hiring, and prioritization decisions Why joining the buyer's journey beats forcing MEDDICC-style sales processes How product-led trials reduce churn and increase win rates What it takes to scale from unicorn to decacorn (and why $100M ARR is the next real milestone)

Northwood Space has raised $100 million in a Series B funding round and announced a new $50 million US Space Force (USSF) contract. NASA has selected Intuitive Machines (IM) as part of 34 global volunteers chosen to track the Artemis II Mission. Blue Origin plans to spend $71.4 million to expand thruster production in Alabama, and more. Remember to leave us a 5-star rating and review in your favorite podcast app. Be sure to follow T-Minus on LinkedIn and Instagram. T-Minus Guest Our guest today is David Buck, Lt. Gen., USAF (Ret.), President, BRPH Mission Solutions. You can connect with David on LinkedIn, and learn more about BRPH on their website. Selected Reading Northwood Space secures a $100M Series B and a $50M Space Force contract- TechCrunch NASA Selects Intuitive Machines to Support Tracking for Artemis II NASA Selects Participants to Track Artemis II Mission Huntsville approves development agreements with Blue Origin and SPX to create more than 450 new jobs MDA Space And Hanwha Sign MOU To Pursue Korean Military Cons NASA Welcomes Oman as Newest Artemis Accords Signatory Europe's First Meteorological Infrared Sounder Reveals the Atmosphere in 3D Share your feedback. What do you think about T-Minus Space Daily? Please take a few minutes to share your thoughts with us by completing our ⁠brief listener survey⁠. Thank you for helping us continue to improve our show.  Want to hear your company in the show? You too can reach the most influential leaders and operators in the industry. Here's our ⁠media kit⁠. Contact us at ⁠space@n2k.com⁠ to request more info. Want to join us for an interview? Please send your pitch to ⁠space-editor@n2k.com⁠ and include your name, affiliation, and topic proposal. T-Minus is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

This is the El Segundo, California-based startup's second raise in less than a year. Also, Ricursive joins Recursive and Unconventional AI in raising massive funds at multi-billion valuations out of the gate. Learn more about your ad choices. Visit podcastchoices.com/adchoices

https://www.youtube.com/watch?v=EF8e-rMB7So .entry-img img{ display:none !important; } .single .hentry .entry-img{ display:none !important; } https://open.spotify.com/episode/3kYIPViq648aqBM8MzHlcM Scaling quickly is every growth company's dream, but it's also the phase where finance is under the greatest threat. Rapid headcount expansion, evolving pricing, complex contracts, and rising investor expectations all hit at once—and every weakness in your finance stack is amplified. Understanding why this phase is so dangerous, and how to design the right controls, systems, and billing infrastructure, is critical if you want to protect cash, avoid revenue leakage, and build a resilient, investor‑ready business. In this episode, Kevin Appleby talks with Shadid Talukder, Strategic Finance Lead at Posh AI, about why the fastest phase of scaling is also the most dangerous for finance. They explore how a lean three-person finance team manages rapid ARR growth, complex enterprise contracts, and investor pressure for both growth and efficiency. Within Posh AI's finance stack, Zenskar plays a central role in billing and revenue recognition for a complex SaaS business selling into banks and credit unions. As pricing and contract structures evolved—across monthly, annual, and multi‑year deals—manual spreadsheets became too risky and operationally heavy. Zenskar now acts as a single system of record for contracts, subscriptions, line items, and future invoices, forecasting and scheduling billing over the life of each deal. This dramatically reduces manual reviews, mitigates missed invoices and revenue leakage, and lets Posh scale billing complexity without proportionally scaling finance headcount or operational risk. Key topics covered: Zero-to-one finance in a fast-scaling AI startup: Shadid joined Posh AI when “the books were literally empty” and helped the company triple ARR while building financial models, reworking an initially non-scalable chart of accounts, and installing core finance processes from scratch  Scaling headcount vs. VC expectations and burn: As Posh grew from ~30 to ~80 FTEs, shifting VC expectations forced a move from “growth at all costs” to tighter burn multiples, proving that rapid scaling without disciplined financial guardrails quickly becomes dangerous for finance  Running a modern finance org with a very lean team: Posh operates with a three-person finance function—SVP Finance, Strategic Finance (Shadid), and Accounting—where no work is “above” anyone, and leaders still handle AP/AR emails themselves, demonstrating what lean but high-caliber finance looks like in practice  Zenskar as a critical control for complex SaaS billing and revenue: To cope with complex, evolving pricing and a mix of monthly, annual, and multi-year contracts, Posh implemented Zenskar as a centralized system of record for contracts, subscriptions, and future invoices—significantly reducing the risk of missed billings and revenue leakage that could materially distort burn and board reporting  Deliberate restraint in tooling and tech stack: After initially “buying software like crazy,” Posh reversed course, cutting underused tools and adopting a strict standard that any new system must have a foundational, clearly justified use case; core stack is QuickBooks + spreadsheets + Zenskar + Ramp, with careful use of GPT for productivity rather than headcount replacement  Balancing growth, profitability, and dilution risk: Shadid outlines that the next phase is defined by sustaining growth while pushing toward profitability, making every incremental hire and dollar of software spend a high-stakes decision—especially when additional fundraising brings dilution, complex cap-table dynamics, and heightened investor pressure for returns About Posh AI Posh AI is an AI‑native SaaS company focused on transforming customer engagement for banks and credit unions. By combining conversational AI with deep domain knowledge of financial services, Posh helps institutions automate routine interactions, deliver personalized experiences, and operate more efficiently, while meeting the strict reliability and compliance standards of regulated industries. About Zenskar   Zenskar is a billing and revenue platform built for modern SaaS companies with complex pricing and contracts. At Posh AI, Zenskar serves as the single source of truth for all customer contracts, subscriptions, and invoice schedules. Once a deal closes, the finance team loads key terms into Zenskar, which then automates invoicing over the contract term. By moving off spreadsheet‑driven billing, Posh AI uses Zenskar to: Reduce manual billing work and one‑off reviews Prevent missed or incorrect invoices that can distort burn and board reporting Confidently support varied billing cadences and sophisticated deal structures This makes Zenskar a core control mechanism that enables Posh to scale faster while keeping finance lean and tightly governed. Links Shadid Talukder on LinkedIn Kevin Appleby on LinkedIn GrowCFO Mentoring Timestamps:  0:00–0:04 Kevin introduces Shadid Talukder and his Strategic Finance role at Posh AI. 0:02–0:04 Shadid shares how he built finance from zero as Posh AI tripled ARR. 0:04–0:06 Posh scaled from ~30 to ~80 FTEs as investor focus shifted to burn efficiency. 0:09–0:11 Posh runs a full finance function with a three-person, hands-on team. 0:11–0:15 Shadid explains why Posh relies on QuickBooks, spreadsheets, and simplicity. 0:15–0:19 Zenskar became the system of record for managing complex SaaS billing and contracts. 0:19–0:23 After overbuying tools, Posh adopted strict controls to keep the stack lean. 0:22–0:23 Custom scripts and APIs replace traditional FP&A platforms. 0:23–0:26 GPT tools are used to boost productivity without adding headcount. 0:27–0:30 Shadid outlines the challenge of growing fast while staying within spend guardrails. 0:30–0:34 The discussion covers Series B trade-offs, dilution, and investor expectations. 0:35–0:38 Shadid reflects on decision pressure and the importance of founder trust. 0:38–0:40 He explains how he operates a high-impact finance role remotely with periodic in-person sessions. Find out more about GrowCFO If you enjoyed this podcast, you can subscribe to the GrowCFO Show with your favorite podcast app. The GrowCFO show is listed in the Apple podcast directory, Spotify and many others. Why not subscribe there today? That way, you never miss an episode. GrowCFO is a great place to extend your professional network. Join GrowCFO as a free member today and participate in our regular networking events and webinars. Premium members can also access our extensive training center and CFO Digital Toolkit. You can enroll in our flagship Future CFO or Finance Leader programs here. You can find out more and join today at growcfo.net

Microsoft granted the FBI access to laptops encrypted with BitLocker. The EU opens an investigation into Grok's creation of sexually explicit images. Glimmers of access pierce Iran's internet blackout. Koi Security warns npm fixes fall short against PackageGate exploits. Some Windows 11 devices fail to boot after installing the January Patch Tuesday updates. CISA warns of active exploitation of  multiple vulnerabilities across widely used enterprise and developer software. ESET researchers have attributed the cyberattack on Poland's energy sector to Russia's Sandworm. This week's business breakdown. Brandon Karpf joins us to talk space and cyber. CISA sits out RSAC.  Remember to leave us a 5-star rating and review in your favorite podcast app. Miss an episode? Sign-up for our daily intelligence roundup, Daily Briefing, and you'll never miss a beat. And be sure to follow CyberWire Daily on LinkedIn. CyberWire Guest Our guest today is cybersecurity executive and friend of the show Brandon Karpf with Dave Bittner and T-Minus Space Daily host Maria Varmazis, for our monthly space and cyber segment. Brandon, Maria and Dave discuss “No more free rides: it's time to pay for space safety.” Selected Reading FBI Accessed Windows Laptops After Microsoft Shared BitLocker Recovery Keys (Hackread) European Commission opens new investigation into X's Grok (The Register) Amid Two-Week Internet Blackout, Some Iranians Are Getting Back Online (New York Times) Hackers can bypass npm's Shai-Hulud defenses via Git dependencies (Bleeping Computer) Microsoft investigates Windows 11 boot failures after January updates (Bleeping Computer) CISA says critical VMware RCE flaw now actively exploited (Bleeping Computer) CISA confirms active exploitation of four enterprise software bugs (Bleeping Computer) ESET Research: Sandworm behind cyberattack on Poland's power grid in late 2025 (ESET)  Aikido secures $60 million in Series B funding. (N2K Pro Business Briefing) CISA won't attend infosec industry's biggest conference (The Register) Share your feedback. What do you think about CyberWire Daily? Please take a few minutes to share your thoughts with us by completing our brief listener survey. Thank you for helping us continue to improve our show.   Want to hear your company in the show? N2K CyberWire helps you reach the industry's most influential leaders and operators, while building visibility, authority, and connectivity across the cybersecurity community. Learn more at sponsor.thecyberwire.com. The CyberWire is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

Multi‑stage AiTM phishing and BEC campaign abusing SharePoint SmarterMail auth bypass flaw now exploited despite patch The problem of AI agents emerges at Davos Huge thanks to our sponsor, Dropzone AI All week we've talked about alert fatigue, MTTR, and the math that's breaking your SOC. Here's the proof. Dropzone AI is trusted by over 300 global enterprises and MSSPs. Named a Gartner Cool Vendor. Recognized in the Fortune Cyber 60. And backed by $37 million in Series B funding. But they're not stopping at a single agent. They're building toward fully agentic SOC teams where human engineers are augmented with specialized AI agents for threat hunting, detection engineering, and forensics. Your team deserves a backup that never sleeps. Book a demo at dropzone.ai. Find the stories behind the headlines at CISOseries.com.

i6 Group is connecting the fragmented aviation fuel ecosystem-airlines, fuel suppliers, and service providers-through a real-time digital platform that eliminates paper-based processes at over 260 airports worldwide. After launching with British Airways at Heathrow in 2015 and recently closing their Series B with German PE firm Itrium, i6 is proving that even heavily regulated, risk-averse industries can achieve step-function operational improvements through software. In this episode of BUILDERS, Alex Mattos, CEO and Managing Director of i6 Group, breaks down how they navigated decade-long enterprise sales cycles, leveraged strategic customers as Series A investors, and are now building toward profitability to maximize exit optionality. Topics Discussed: The surprising analog nature of aviation fuel operations despite advanced aircraft technology i6's pivot from defense fuel system testing to commercial aviation digitization The multi-party fuel ecosystem: airlines, suppliers, service providers, and logistics chains Strategic approach to landing British Airways and Virgin Atlantic as launch customers Fundamental differences between European fuel optimization vs. US supply chain management models Multi-stakeholder enterprise sales involving fuel teams, flight ops, pilot unions, and CFOs Strategic Series A with customer-investors: British Airways, JetBlue, Shell, and World Fuel Services Series B transition from strategic to PE backing focused on scaling operations and go-to-market Network effects driving compounding value as airport coverage expands Path to self-sustainability and exit strategy considerations GTM Lessons For B2B Founders: Target brand DNA, not just budget, for early enterprise customers: i6 deliberately approached Virgin Atlantic because of Richard Branson's reputation for "being entrepreneurial, taking a risk, doing something different." This wasn't naive brand worship—it was strategic targeting based on organizational risk tolerance. When selling complex infrastructure to enterprises pre-product-market fit, a prospect's innovation track record matters more than their budget size. Map your early pipeline based on cultural willingness to partner with startups, not just technical fit. Invest in non-paying reference customers as currency for tier-one deals: Virgin Atlantic became i6's first operational deployment without payment. This wasn't charity—it was strategic capital allocation. The working reference at Virgin directly unlocked British Airways: "we turned up, demonstrated what we were doing...we've done this trial with Virgin and here's the results, and it went really well." For founders selling to conservative enterprises, one live deployment at a credible brand is worth more than a dozen pitch decks. Budget 6-12 months of runway for strategic pilots that generate proof points, not revenue. Create forcing functions with specific follow-up commitments: When British Airways said "if you're still here in six months, come back," most founders would hear soft rejection. Alex heard a calendar commitment and returned "to the day" with results. This precision signaling—we take your requirements seriously enough to track them to the day—separates serious vendors from tire-kickers. When enterprises set conditional bars, treat them as binding contracts and demonstrate execution discipline through exact follow-through. Position for market disruption by maintaining warm enterprise relationships: i6 benefited when an incumbent competitor liquidated, creating urgent procurement needs at British Airways. But luck favors the prepared—they had already established credibility through their Virgin deployment. Maintain enterprise relationships even when deals seem stalled. In concentrated B2B markets, competitive exits, budget releases, and trigger events happen regularly. Your position in the consideration set when disruption hits determines whether you capture the opportunity. Engineer word-of-mouth in concentrated industries through excellence, not marketing: Four months after Heathrow deployment, Dubai airport approached i6 unsolicited: "we've heard great things." In the aviation fuel community—which Alex describes as "surprisingly small"—exceptional execution travels faster than any outbound motion. This changes GTM strategy: in concentrated industries, over-invest in customer success and operational excellence at early deployments rather than spreading thin across many accounts. Your first customers are your sales team. Segment GTM by operational model, not just geography or company size: i6 discovered European airlines optimize for fuel efficiency and real-time decisions, while US airlines (controlling their own supply networks since the late 1980s) prioritize supply chain visibility: "how much fuel did we put in the plane, how much have we had delivered, how much have we got left." These aren't feature preferences—they're fundamentally different jobs-to-be-done driven by market structure. Don't assume global enterprises have unified needs. Segment by operational model and regulatory environment, then customize messaging and roadmap accordingly. Stage investor expertise to match company evolution, not just valuation milestones: Series A brought strategic investors who were actual users (British Airways, JetBlue, Shell, World Fuel Services) for product validation and network access. Series B brought PE firm Itrium for "scaling the business...building and growing our sales and revenue teams." This wasn't opportunistic—it was deliberate staging of capital sources to match capability gaps. Don't optimize fundraising purely on valuation or dilution. Map your next 18-month bottleneck (product validation vs. operational scaling vs. market expansion) and raise from investors who've solved that specific problem. Build for profitability to control your exit timing and terms: Alex's goal is avoiding Series C entirely: "we build and establish a fully self-sustaining business...the business becomes fully sustainable in the next couple of years." This isn't conservatism—it's strategic optionality. Reaching profitability eliminates the forced march toward subsequent rounds, letting you choose between IPO or M&A based on market conditions rather than cash position. For infrastructure plays with long implementation cycles, factor sustainability into your growth model early, even if it moderates topline growth rates. // Sponsors: Front Lines — We help B2B tech companies launch, manage, and grow podcasts that drive demand, awareness, and thought leadership. www.FrontLines.io The Global Talent Co. — We help tech startups find, vet, hire, pay, and retain amazing marketing talent that costs 50-70% less than the US & Europe. www.GlobalTalent.co // Don't Miss: New Podcast Series — How I Hire Senior GTM leaders share the tactical hiring frameworks they use to build winning revenue teams. Hosted by Andy Mowat, who scaled 4 unicorns from $10M to $100M+ ARR and launched Whispered to help executives find their next role. Subscribe here: https://open.spotify.com/show/53yCHlPfLSMFimtv0riPyM

For the last eight years, I've publicly shared my conviction around “relaxation” building into the next functional CPG frontier, due to the growing consumer demand from today's overstimulated (especially younger) generations for products that enhance mental wellness, support relaxation and stress relief, and enable alcohol moderation. Also, during that same timeframe…I've highlighted only one single brand (repeatedly) which I believed could become the “Red Bull of Relaxation,” effectively pioneering a category counterbalancing the $26 billion U.S. energy drinks market built on stimulation. “Take a Recess.” And it might be corny to make this comparison but hearing that Recess brand tagline was like a Jerry McGuire “you had me at hello” moment. Regardless, it became super apparent to me that founder (Ben Witte) truly understood Recess would only have a chance at becoming the definitive household name in modern relaxation if the selling formula started with emotion. Obviously, there's A LOT of other internal/external business dynamics ultimately at play…and the Recess story hasn't been without twists, turns, and challenges, but recently it's entering a fundamentally new chapter. So, I was honored when (right after) its $30 million Series B fundraising (and associated leadership hiring) news was released…I got a text message from Ben Witte asking if I'd be interested in hosting himself and Kyle Thomas for their first official recorded Co-CEO fireside chat together. As you'd imagine, in an effort to best help them share the nuanced business story of how Recess is scaling into the next iconic modern beverage company…it required a wide-reaching strategic conversation, but one that undoubtedly will provide insightful nuggets across every corner of the CPG industry.

This week's throwback guest, Rob Zambito, went from being a shy child, to University of Penn Psychology graduate, to now a successful Customer Success Leader & Consultant. Rob has experience from Seed to Series B, growing Customer Success teams from 2 to 20 members, and so much more!In this week's episode, we discussed:Building A Common Language Across The BusinessA Hypothesis-Driven ApproachThe Importance of MentorsOnboarding Cross-Sell Checklist Consciously Investing in Customer Success Much MorePlease enjoy this week's episode with Rob Zambito ____________________________________________________________________________I am now in the early stages of writing my first book! In this book, I will be telling my story of getting into sales and the lessons I have learned so far, and intertwine stories, tips, and advice from the Top Sales Professionals In The World! As a first time author, I want to share these interviews with you all, and take you on this book writing journey with me! Like the show? Subscribe to the email: https://mailchi.mp/a71e58dacffb/welcome-to-the-20-podcast-communityI want your feedback!

In this episode of The Med-Tech Talent Lab, host Mitch Robbins sits down with Jahnavi Lokre, Chief Operating Officer at Diality, a Series B medical device company focused on transforming kidney care.Jahnavi shares her 25+ year journey from growing up in India in a science-driven household to building a career across software engineering, regulated industries, and executive leadership in MedTech. With experience spanning transportation systems, contract manufacturing, and product ownership at both large organizations and startups, she offers a rare, cross-industry perspective on what it truly takes to build and scale high-performing teams.Listeners will gain practical insight into:How transferable skills from highly regulated industries can unlock MedTech leadership opportunitiesWhy stepping up before you feel “ready” is often the catalyst for executive growthThe mindset shifts required to move from engineering into senior leadershipHiring lessons learned the hard way, including when to act quickly on a mis-hireHow strong leaders balance trust, accountability, and executionWhy bold thinking, persistence, and values-driven leadership matter in fast-moving startupsJahnavi also discusses Diality's mission to improve outcomes and reduce costs in dialysis care, the realities of innovating in an under-evolved market, and what excites her most about the future of kidney disease treatment.This episode is packed with candid leadership lessons, career-defining insights, and actionable advice for engineers, operators, and aspiring executives looking to grow their impact in MedTech and beyond.

"We want musicians to keep making music, and for fans to have confidence that the music they find on Bandcamp was created by humans," the company said. Also, Type One Energy has raised more than $160 million from investors to date. The new funding will help the company further refine its stellarator technology. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Amias Gerety, Partner at QED Investors, brings an unconventional perspective to venture capital shaped by his eight years at the US Treasury Department during the financial crisis. A mechanical thinker, Amias applies an essentialist approach to understanding how businesses work. He explains why QED looks for companies that triple every six months at Series A, how inverted AI creates new opportunities in financial services, and why the best advice for founders remains timeless: build something people want and charge more than it costs to make. With insights on the AI bubble, the application layer renaissance, and why saying no 99 times out of 100 is the real job of a VC, Amias offers a masterclass in disciplined, thesis-driven investing.In this episode, you'll learn:[01:24] Amias's unique path from politics and Treasury to venture capital[05:13] The lever theory: how government and VC create systemic change[07:12] Why mechanical thinking and first principles matter in VC[14:48] QED's investment sweet spot: Series A and series B with undeniable momentum[19:25] What product-market fit really means and how to recognize it[22:14] Inverted AI: Why the world needs financial services for the AI economy[26:43] The AI bubble paradox: overvalued companies, transformative technology[32:57] Why early-stage founders should ignore the macro and focus on customers[34:31] The brutal math of ventureThe nonprofit organization Amias is passionate about: EastersealsAbout Amias GeretyAmias Gerety is a Partner at QED Investors, where he focuses on FinTech and InsurTech investments. Before joining QED in 2017, Amias spent eight years at the US Treasury Department from the first day of the Obama administration through its final day. During his tenure, he helped write the Dodd-Frank Act and built the Financial Stability Oversight Council, the organization responsible for monitoring systemic risk in the US financial system. His government experience during the financial crisis gives him a unique perspective on market dynamics and regulatory frameworks. A mechanical thinker who approaches investments with an essentialist mindset, Amias has invested in companies like Kin Insurance, Prosper, and Tint. He previously worked as a management consultant and with Save the Children in East Africa.About QED InvestorsQED Investors is one of the most successful venture capital firms focused on FinTech investments globally. As a multi-stage, global firm with a $650 million early-stage fund and $300 million growth fund, QED specializes in Series A and B investments in companies demonstrating exceptional momentum and product-market fit. The firm requires portfolio companies to show dramatic growth—expecting tripling in six months for Series A and tripling in a year for Series B investments. QED's partners bring deep domain expertise from building and scaling financial services companies, with a particular focus on companies that are reshaping financial services through technology. The firm is known for its rigorous, thesis-driven approach to investing and its high conviction in backing founders who have found authentic product-market fit in large, expanding markets.Subscribe to our podcast and stay tuned for our next episode.

The FBI warns of Kimsuky quishing. Singapore warns of a critical vulnerability in Advantech IoT management platforms. Russia's Fancy Bear targets energy research, defense collaboration, and government communications. Malaysia and Indonesia suspend access to X. Researchers warn a large-scale fraud operation is using AI-generated personas to trap mobile users in a social engineering scam. BreachForums gets breached. The NSA names a new Deputy Director. Monday Biz Brief. Our guest is Sasha Ingber, host of the International Spy Museum's SpyCast podcast. The commuter who hacked his scooter.  Remember to leave us a 5-star rating and review in your favorite podcast app. Miss an episode? Sign-up for our daily intelligence roundup, Daily Briefing, and you'll never miss a beat. And be sure to follow CyberWire Daily on LinkedIn. CyberWire Guest Today we are joined by Sasha Ingber, host of the International Spy Museum's SpyCast podcast, on the return of SpyCast to the N2K CyberWire network. Selected Reading North Korea–linked APT Kimsuky behind quishing attacks, FBI warns (Security Affairs)  Advantech patches maximum-severity SQL injection flaw in IoT products (Beyond Machines) Russia's APT28 Targeting Energy Research, Defense Collaboration Entities (SecurityWeek) Malaysia and Indonesia block X over deepfake smut (The Register) New OPCOPRO Scam Uses AI and Fake WhatsApp Groups to Defraud Victim (Hackread) BreachForums hacking forum database leaked, exposing 324,000 accounts (Bleeping Computer) Former NSA insider Kosiba brought back as spy agency's No. 2 (The Record) Vega raises $120 million in a Series B round led by Accel. Reverse engineering my cloud-connected e-scooter and finding the master key to unlock all scooters (Rasmus Moorats) Share your feedback. What do you think about CyberWire Daily? Please take a few minutes to share your thoughts with us by completing our brief listener survey. Thank you for helping us continue to improve our show. Want to hear your company in the show? N2K CyberWire helps you reach the industry's most influential leaders and operators, while building visibility, authority, and connectivity across the cybersecurity community. Learn more at sponsor.thecyberwire.com. The CyberWire is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

In this episode of Run the Numbers, CJ sits down with Jason Kong, General Partner at Base10 Ventures, to unpack the firm's focus on “automation for the real economy” — software built for industries most tech investors overlook, but the world depends on. Jason breaks down what makes Series B investing uniquely hard, how he evaluates back-office and vertical SaaS opportunities, and where markets tip from niche to overcrowded. They also discuss Base10's decision to donate 50% of profits to fund scholarships, plus a lightning round spanning fantasy football, shorting SaaS in 2022, and a venture take that might spark debate.—SPONSORS:Metronome is real-time billing built for modern software companies. Metronome turns raw usage events into accurate invoices, gives customers bills they actually understand, and keeps finance, product, and engineering perfectly in sync. That's why category-defining companies like OpenAI and Anthropic trust Metronome to power usage-based pricing and enterprise contracts at scale. Focus on your product — not your billing. Learn more and get started at https://www.metronome.comRightRev is an automated revenue recognition platform built for modern pricing models like usage-based pricing, bundles, and mid-cycle upgrades. RightRev lets companies scale monetization without slowing down close or compliance. For RevRec that keeps growth moving, visit https://www.rightrev.comRillet is an AI-native ERP built for modern finance teams that want to close faster without fighting legacy systems. Designed to support complex revenue recognition, multi-entity operations, and real-time reporting, Rillet helps teams achieve a true zero-day close—with some customers closing in hours, not days. If you're scaling on an ERP that wasn't built in the 90s, book a demo at https://www.rillet.com/cjTabs is an AI-native revenue platform that unifies billing, collections, and revenue recognition for companies running usage-based or complex contracts. By bringing together ERP, CRM, and real product usage data into a single system of record, Tabs eliminates manual reconciliations and speeds up close and cash collection. Companies like Cortex, Statsig, and Cursor trust Tabs to scale revenue efficiently. Learn more at https://www.tabs.com/runAbacum is a modern FP&A platform built by former CFOs to replace slow, consultant-heavy planning tools. With self-service integrations and AI-powered workflows for forecasting, variance analysis, and scenario modeling, Abacum helps finance teams scale without becoming software admins. Trusted by teams at Strava, Replit, and JG Wentworth—learn more at https://www.abacum.aiBrex is an intelligent finance platform that combines corporate cards, built-in expense management, and AI agents to eliminate manual finance work. By automating expense reviews and reconciliations, Brex gives CFOs more time for the high-impact work that drives growth. Join 35,000+ companies like Anthropic, Coinbase, and DoorDash at https://www.brex.com/metrics—LINKS:Jason on LinkedIn: https://www.linkedin.com/in/jasonykong/Base10 Partners: https://base10.vc/CJ on LinkedIn: https://www.linkedin.com/in/cj-gustafson-13140948/Mostly metrics: https://www.mostlymetrics.com—RELATED EPISODES:Scaling to $1B+ Revenue: From ServiceNow to Samsara | Dominic Phillipshttps://youtu.be/vBY6WZBMljw—TIMESTAMPS:00:00:00 Preview and Intro00:02:20 Sponsors — Metronome, RightRev, Rillet00:06:02 Base10 Background00:06:41 Automation for the Real Economy00:09:27 Vertical vs. Horizontal Software00:10:38 Cash Flow and Durability00:11:19 Product-Market Fit and ROI00:12:56 Growth Limits Selling to Tech00:13:19 The Size of the Real Economy00:14:16 Sponsors — Tabs, Abacum, Brex00:18:50 Base10's Giving Model00:20:30 Access, Education, and Tech00:21:53 Purpose and Founder Alignment00:22:51 Radical Transparency00:23:56 Portfolio Focus and Strategy00:24:05 Investing Ahead of Consensus00:26:29 ERP Adjacency as Alpha00:28:58 Lessons From Hedge Funds00:32:29 Public Markets Reality00:34:05 Public vs. Private Investing00:34:48 The Series B Sweet Spot00:36:49 A Bifurcated Series B Market00:38:56 Fast Series Bs and 2021 Vibes00:42:16 What Series B Looks Like Now00:44:36 Back Office Automation00:46:02 ERP-Centric Workflows00:48:33 Long-Ass Lightning Round00:49:36 Shorting SaaS in 202200:50:16 Fantasy Football and Investing00:52:57 Career Advice That Surprises00:55:03 A Contrarian Venture Take00:56:22 Credits#RunTheNumbersPodcast #SeriesB #RealEconomy #VerticalSaaS #BackOfficeAutomation This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit cjgustafson.substack.com

In this episode of The SaaS Revolution Show, Alex Theuma is joined by Nick Turner, CEO of Dreamdata, to discuss the journey from CRO to CEO and what it really takes to scale a B2B SaaS company in the age of AI. Nick shares lessons from Dreamdata's growth journey, including the company's $55M Series B, and explains why trust and accuracy matter more than hype when building AI products. He breaks down the risks of applying generative AI and agents to complex revenue and attribution data and what SaaS leaders should consider before putting AI in front of customers, boards, or finance teams. Alex and Nick also discuss: - Nick's transition from CRO to CEO and what changed at the leadership level. - How Dreamdata approaches AI as a system of context, not just automation. - Why reliable attribution and data integrity are critical for modern GTM teams. - How investors evaluate AI, retention, and fundamentals at growth stage. - Practical advice for founders building sustainable, predictable SaaS businesses into 2026.       Check out the other ways SaaStock is helping SaaS founders move their business forward: 

Read a transcript of this episode. Subscribe to receive transcripts.What we call waste is really just misallocated feedstock—raw materials waiting to be cycled back into the next generation of products and packaging. According to research by the World Economic Forum and United Nations Development Programme, the circular economy could unlock $4.5 trillion in new global value by 2030, and investors are racing to capture part of that opportunity. Meet Elizabeth Blankenship-Singh, Director of Innovation at Overlay Capital, an Atlanta-based alternative investment firm whose Waste and Materials Fund is backing both early-stage materials innovators and later-stage recycling operations with established infrastructure. Overlay's strategy involves investing in innovation and implementation simultaneously—in both startups and established companies—to accelerate progress across multiple layers of the circular economy. It offers a window into where smart money sees the materials transition heading. Elizabeth explains that sortation is the biggest bottleneck at the materials recycling facilities (MRFs) your garbage and recycling are sent to after curbside collection. The U.S. is simultaneously the world's leading exporter of scrap aluminum and the number one importer of finished aluminum, because we've lacked domestic sorting capacity. Overlay has invested in companies like AMP Robotics, which recently closed a 20-year contract with SPSA, a southeastern Virginia municipal authority, to sort all recyclables from four to five cities using AI-driven systems. When you fix sortation, she says, you trigger a domino effect: recycling rates climb, landfill life extends, and margins improve as higher-purity materials command premium prices. Overlay's portfolio also includes next-generation materials companies united by a common thesis: they must be better, faster, cheaper, and more sustainable than what they replace. Cruz Foam converts chitin from shrimp shells into compostable packaging foam. Simplifyber uses cellulose to create biodegradable soft goods through 3D molding, bypassing traditional textile manufacturing entirely. Terra CO2 just closed a $124 million Series B to scale low-carbon cement technology that could cut into concrete's 8% share of annual global CO2 emissions. Each uses abundant, waste-derived feedstocks and has achieved or is on a clear path to price parity with incumbents.You can learn more about Overlay Capital at overlaycapital.comSubscribe to Sustainability In Your Ear on iTunesFollow Sustainability In Your Ear on Spreaker, iHeartRadio, or YouTube

Today on the ThinkData Podcast, I'm joined by Arjun Vora, Co-Founder of Teambridge, a Series B startup redefining how frontline and hourly teams are managed.Arjun previously led design on the Uber for Drivers app, an experience that gave him first-hand insight into how broken tools, poor communication, and rigid systems impact people who don't sit at desks all day.Teambridge is building a single operating system for frontline teams: combining scheduling, communication, workflows, and real-time insights into one intelligent platform. Their mission is simple but powerful, make managing people easier and make work better for the people doing it.In this episode, we explore why frontline workers have been overlooked by technology, how AI can genuinely improve shift-based work, and where the biggest opportunities lie over the next few years.

Welcome to the Personal Development Trailblazers Podcast! In today's episode, we're talking about how to intentionally design a life you love, starting from the inside out.Rachel Korb is an ICF Professional Certified Coach and founder of Nervous System Mastery for Women. She's delivered 800+ coaching hours across 41 nationalities and spent over a decade scaling remote startups from seed to Series B across 36 countries.She started practicing mindfulness and somatics at 19, yet still burned out three times in her startup career. The third time—marked by an intestinal bleed, insomnia, and a shingles outbreak she initially mistook for a jellyfish sting—was the turning point. She realized she'd been forcing a 28-day cycle into a 24-hour framework. Applying male-patterned productivity to female physiology—and wondering why she kept crashing.Now she teaches high-achieving women to work with their biology instead of against it. Women's nervous systems function differently—capacity adjusts with the monthly cycle, not just the circadian rhythm.Her mission goes beyond individual transformation: when women stop accepting systems designed to exhaust them, they shift company cultures and challenge the paradigm that says burnout is the price of success. Her message? There's nothing wrong with you. You're just working with the wrong framework.Connect with Rachel Here: https://www.linkedin.com/in/rachelkorb/https://coachingwithrachel.substack.com/https://rachelkorb.com/Grab the freebie here: Calm Self-assessment: https://rachelkorb.com/free-resources===================================If you enjoyed this episode, remember to hit the like button and subscribe. Then share this episode with your friends.Thanks for watching the Personal Development Trailblazers Podcast. This podcast is part of the Digital Trailblazer family of podcasts. To learn more about Digital Trailblazer and what we do to help entrepreneurs, go to DigitalTrailblazer.com.Are you a coach, consultant, expert, or online course creator? Then we'd love to invite you to our FREE Facebook Group where you can learn the best strategies to land more high-ticket clients and customers. QUICK LINKS: APPLY TO BE FEATURED: https://app.digitaltrailblazer.com/podcast-guest-applicationDIGITAL TRAILBLAZER: https://digitaltrailblazer.com/

In this episode, Sasha Orloff sits down with Sam Ross, founder and CEO of Numeral and former product leader at Teespring and Airbnb, about raising Series B funding from Mayfield (following Benchmark's Series A and Uncork's seed) to build what he calls "the most boring AI company"—an end-to-end sales tax automation platform that uses AI to eliminate the manual burden of multi-state and international tax compliance for e-commerce and SaaS businesses, transforming a traditionally services-heavy industry into a fully automated solution that handles everything from nexus analysis and registration to filings and government correspondence. -- SPONSORS: Notion Boost your startup with Notion—the ultimate connected workspace trusted by thousands worldwide! From engineering specs to onboarding and fundraising, Notion keeps your team organized and efficient. For a limited time, get 6 months of Notion AI FREE to supercharge your workflow. Claim your offer now at ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://notion.com/startups/puzzle⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Puzzle

What if the most powerful communicators aren't the most polished ones? In this episode, Dia Bondi shares how watching a three-day communication skills class transformed her life. She had no resume to support her desire, but she made the ask that changed everything: "Will you apprentice me?" That mentor said yes, and 25 years later, she's still helping leaders speak powerfully from who they truly are. Dia's work goes beyond messaging to something deeper: helping clients articulate what they have to say before figuring out how to say it. She's coached VC-backed founders past Series B, VP-level executives, and even helped Rio de Janeiro win the right to host the 2016 Olympics (a $7 billion decision) without ever knowing who referred her. Dia reveals the six attributes of the most compelling communicators: they make the choice to be big in big moments, they know how to make strategic asks, they absolutely know their voice, they've got killer setups for their asks, they bridge their voice to the business really well, and they prep on purpose by actually talking through what they'll say.  She shares the story of a founder who created a 40-minute story to invite existing customers to help inform what gets built next. That task drove $10 million of value in the business. Dia also explains why the most powerful moment in her career might have been her first deal that included both cash and shares, because once she did it once, she now had the confidence and understanding to recreate it over and over again.   [00:08:20] The Ask That Changed My Life Watched three-day communication skills class about storytelling in business Had no resume to support desire to do that work Asked if he would apprentice her, and he did 25 years later still finds it compelling because it truly moves the needle for people [00:10:20] What You Have to Say vs. How to Say It Messaging is "how do I say it?" but Dia lives in the step before that "What is it that you have to say?" Until a leader can articulate things that feel innate but aren't obvious, can't create the message First articulate what you have to say, then figure out how best to express it [00:12:00] The 40-Minute Story Worth $10 Million Former client needed to invite existing customers to participate in informing product development Put together compelling story bridging what founder has to say with story of business Set up really killer ask, 40-minute story to set up that ask Client secured participation needed, will drive about $10 million of value in business [00:17:20] The First Deal With Shares Introduction to blockchain company working directly with founders First time setting up deal that included both cash and shares It paid off, and importantly gave confidence and understanding Once you do it first time, now have opportunity to do it over and over because it's in your radar [00:20:00] Helping Rio Win the Olympics bid Got a request over the internet: "We heard you're good, can you give us a call?" Turned out to be production team working on Rio's bid for 2016 Olympics $7 billion decision, got to work in those teeny tiny rooms where stories get crafted Still doesn't know who referred her, but knows what part of work it came from [00:24:40] Six Attributes of Powerful Communicators One: Make choice to be big in big moments (not about being polished or showboat) Two: Know how to make strategic asks and use asking strategically Three: Absolutely know their voice and can give good feedback to creative collaborators Four: Got killer setups for the asks they make of partners [00:27:20] The Last Two Attributes Five: Bridge their own voice to the business really well Six: Prep on purpose by actually talking through it, not just thinking about it One founder's pregame routine: end call, go downstairs, change shirt, wash face, drink water If he did that six times in day, he'd go through six shirts [00:30:00] Kevin's $16.1 Million Conversation 2003, owned cleaning and restoration company with website producing $15,000/month Mentor Joe Polish: "Come speak at event, document what you do into a course" Had never spoken publicly before, invested in Speaker's Bootcamp training Made $35,000 in course sales that day, but that wasn't the big win [00:32:40] The Real Big Win Over next 12 years sold $16.1 million worth of that course All because of one conversation: "Why don't you come down and speak?" Brought energy, joyfulness, willingness to share with generosity When you're invited, say yes   KEY QUOTES "I made the ask that changed my life. That gentleman brought me in and let me watch his class. I asked if he would apprentice me, and he did." - Dia Bondi "The most compelling communicators make the choice to be big in big moments. And that's never about being a showboat." - Dia Bondi CONNECT WITH DIA BONDI 

NATO suspects Russia is developing a new anti-satellite weapon to disrupt the Starlink network. A failed polygraph sparks a DHS probe and deepens turmoil at CISA. A look back at Trump's cyber policy shifts. MacSync Stealer adopts a stealthy new delivery method.  Researchers warn a popular open-source server monitoring tool is being abused. Cyber criminals are increasingly bypassing technical defenses by recruiting insiders. Scripted Sparrow sends millions of BEC emails each month. Federal prosecutors take down a global fake ID marketplace. Monday business brief. Our guest is Eric Woodruff, Chief Identity Architect at Semperis, discussing "NoAuth Abuse Alert: Full Account Takeover." Atomic precision meets Colorado weather. Remember to leave us a 5-star rating and review in your favorite podcast app. Miss an episode? Sign-up for our daily intelligence roundup, Daily Briefing, and you'll never miss a beat. And be sure to follow CyberWire Daily on LinkedIn. CyberWire Guest Today on our Industry Voices, we are joined by Eric Woodruff, Chief Identity Architect at Semperis, discussing "NoAuth Abuse Alert: Full Account Takeover." Tune into the full conversation here. Selected Reading Starlink in the crosshairs: How Russia could attack Elon Musk's conquering of space (AP News) Project West Ford (Wikipedia) Acting CISA director failed a polygraph. Career staff are now under investigation (POLITICO) Dismantling Defenses: Trump 2.0 Cyber Year in Review (Krebs on Security) MacSync macOS Malware Distributed via Signed Swift Application (SecurityWeek) From ClickFix to code signed: the quiet shift of MacSync Stealer malware (Jamf)  Hackers Abuse Popular Monitoring Tool Nezha as a Stealth Trojan (Hackread) Cyber Criminals Are Recruiting Insiders in Banks, Telecoms, and Tech (Check Point) Scripted Sparrow Sends Millions of BEC Emails Each Month (Infosecurity Magazine) FBI Seizes Fake ID Template Domains Operating from Bangladesh (Hackread) Adaptive Security raises $81 million in a Series B round led by Bain Capital Ventures. (N2K Pro) NIST tried to pull the pin on NTP servers after blackout caused atomic clock drift (The Register) Share your feedback. What do you think about CyberWire Daily? Please take a few minutes to share your thoughts with us by completing our brief listener survey. Thank you for helping us continue to improve our show. Want to hear your company in the show? N2K CyberWire helps you reach the industry's most influential leaders and operators, while building visibility, authority, and connectivity across the cybersecurity community. Learn more at sponsor.thecyberwire.com. The CyberWire is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

This week on The Data Minute, Peter sits down with Ben Orthlieb, Founding Partner at Blue Moon VC, for a look under the hood of a firm that has completely re-engineered the venture capital process using AI.Ben explains how Blue Moon uses a proprietary tech stack to source over 12,000 teams a year and screen them down to the top 5% based on their probability of graduating to Series B, achieving performance metrics that rival top-tier firms without the massive headcount. He breaks down why the "warm intro" is obsolete, how sending AI-generated dossiers to founders results in a 75% meeting acceptance rate, and why human judgment is still the final decision-maker.They also discuss the "hollowing middle" of the venture market, why multi-billion dollar funds struggle to innovate their own workflows, and how a small check strategy allows Blue Moon to cooperate, rather than compete, with the biggest names in the industry.Subscribe to Carta's weekly Data Minute newsletter: https://carta.com/subscribe/data-newsletter-sign-up/Explore interactive startup and VC data, with Carta's Data Desk: https://carta.com/data-desk/Chapters00:00 – Intro: The AI-first VC01:27 – Blue Moon's thesis: Coverage and Winning03:11 – How to source 12,000 teams a year without a network05:40 – "Machine learning instinct": Optimizing for Series B graduation07:44 – Backtesting the algorithm against top 50 Seed firms10:12 – The 75% meeting conversion rate (and why cold email works)12:30 – The "AI Dossier": Showing founders you did the work14:13 – Finding outliers outside the "Credibility Pool" (The Mercor story)16:05 – The investment process: Where AI ends and humans begin18:43 – Does it matter who else is on the cap table?23:36 – The "Small Check" advantage in winning allocation25:22 – How to interview for resilience30:20 – Why personal questions are a competitive advantage32:31 – Follow-ons, reserves, and systematic secondaries36:00 – Why haven't big funds copied this strategy yet?39:54 – The "hollowing middle" of the VC market44:40 – Why brand is the only defense against noise49:20 – Do warm intros actually result in better investments?52:46 – The future of the "Operator-VC" model56:00 – What LPs really think about an AI-driven fund59:07 – OutroThis presentation contains general information only and eShares, Inc. dba Carta, Inc. (“Carta”) is not, by means of this publication, rendering accounting, business, financial, investment, legal, tax, or other professional advice or services, and is for informational purposes only.  This presentation is not a substitute for such professional advice or services nor should it be used as a basis for any decision or action that may affect your business or interests. © 2025 eShares, Inc., dba Carta, Inc. All rights reserved.

Dexory builds data intelligence platforms for logistics, using autonomous robots to create digital twins of warehouse operations. With over $280 million raised through a recent preemptive Series C, the company has scaled from a bootstrapped startup to a full-stack robotics operation expanding across Europe and the US. In this episode of Category Visionaries, I sat down with Andrei Danescu, Founder and CEO of Dexory, to unpack how the company navigated early product-market misalignment, cracked the messaging for a category-creating technology, and maintained execution velocity as a capital-intensive business. Topics Discussed: Building in logistics after observing parts tracking failures in Formula One operations The costly mistake: spending years on public space robots before committing to warehouse logistics Why bootstrapping for five to six years forced product discipline before venture funding Messaging shift from autonomous robot capabilities to inventory visibility pain points Zero infrastructure change as a strategic product constraint for live warehouse deployments Geographic expansion strategy using multinational customers for internal reference selling How the convergence of AI adoption, sensor cost reduction, and industry data appetite created market timing Maintaining commercial velocity as the primary metric for Series C readiness in full-stack businesses GTM Lessons For B2B Founders: Message to the problem, not the technology stack: When Dexory led with "world's tallest autonomous robots" and "scan 10,000+ pallets per hour," prospects responded with "what does it actually do?" The shift to leading with inventory visibility and stock control—a pain point customers immediately recognized—unlocked early traction. For category-creating products, customers need to map your solution to existing problems before they can appreciate technical differentiation. Andrei's insight: start with the problem customers know they have, then layer in technical superiority once you've established relevance. Turn operational constraints into product requirements: Dexory designed around the reality that warehouses operate as "live businesses" that cannot pause for infrastructure overhauls. Zero infrastructure change became a core product spec, not a nice-to-have feature. This required autonomous navigation in complex, dynamic environments rather than controlled spaces. Founders building for established industries should identify non-negotiable operational constraints early and architect solutions that respect them rather than requiring customers to adapt their operations. Build value expansion mechanisms before closing your first customer: Dexory established infrastructure for continuous product improvement from day one, treating early deployments as ongoing collaborations rather than transactions. Customers influenced roadmap priorities while Dexory delivered incremental value increases over time. This transformed buyers into advocates who took "point of pride" in the technology. The tactical approach: structure customer agreements and product architecture to support continuous delivery cycles that compound value rather than one-time implementations. Use multinational customers as geographic expansion infrastructure: Instead of opening regional offices across territories, Dexory targeted global companies where a European deployment could generate US interest through internal reference calls. Andrei noted this creates "a lot stronger" references "because they're already part of the same company." The expansion velocity this enabled—UK to Europe to US without massive regional buildout—proved critical for a capital-intensive business. Founders should prioritize customers with multi-region operations who can accelerate geographic reach through internal advocacy networks. Treat post-raise execution velocity as your next round metric: After Dexory's Series B, investors returned a month later to find the company "already ahead of plan." This consistent over-delivery on growth targets set up their preemptive Series C. For full-stack businesses where each dollar deployed takes longer to show returns, maintaining commercial momentum signals execution capability that justifies higher valuations. Andrei's warning: the temptation to slow down and "invest a bit more in product" after raising capital is exactly when founders need to double down on commercial traction as the North Star. // Sponsors: Front Lines — We help B2B tech companies launch, manage, and grow podcasts that drive demand, awareness, and thought leadership. www.FrontLines.io The Global Talent Co. — We help tech startups find, vet, hire, pay, and retain amazing marketing talent that costs 50-70% less than the US & Europe. www.GlobalTalent.co // Don't Miss: New Podcast Series — How I Hire Senior GTM leaders share the tactical hiring frameworks they use to build winning revenue teams. Hosted by Andy Mowat, who scaled 4 unicorns from $10M to $100M+ ARR and launched Whispered to help executives find their next role. Subscribe here: https://open.spotify.com/show/53yCHlPfLSMFimtv0riPyM

Apple and Google issue emergency updates to patch zero-days.  Google links five additional Chinese state-backed hacking groups to “React2Shell.” France's Ministry of the Interior was hit by a cyberattack. Atlassian patches roughly 30 third-party vulnerabilities. Microsoft says its December 2025 Patch Tuesday updates are breaking Message Queuing. Researchers uncovered a massive exposed database with nearly 4.3 billion professional records openly accessible online. Britain's new MI6 chief warns of an “aggressive, expansionist, and revisionist” Russia. Monday Business Brief. On today's Threat Vector, ⁠Michael Heller⁠ from Unit 42 chats with security leaders ⁠Greg Conti⁠ and ⁠Tom Cross⁠ to unpack the hacker mindset and the idea of “dark capabilities”. A cyber holiday gift guide for the rest of us.  Remember to leave us a 5-star rating and review in your favorite podcast app. Miss an episode? Sign-up for our daily intelligence roundup, Daily Briefing, and you'll never miss a beat. And be sure to follow CyberWire Daily on LinkedIn. Threat Vector Segment In this segment of Threat Vector, host ⁠Michael Heller⁠, Managing Editor for Cortex and Unit 42 and Executive Producer of the podcast, sits down with long-time security leaders ⁠Greg Conti⁠ and ⁠Tom Cross⁠ to unpack the hacker mindset and the idea of “dark capabilities” inside modern technology companies. You can listen to their full discussion here. Be sure to catch new episodes of Threat Vector by Palo Alto Networks every Thursday on your favorite podcast app. Selected Reading Apple, Google forced to issue emergency 0-day patches (The Register) Google links more Chinese hacking groups to React2Shell attacks (Bleeping Computer) French Interior Ministry confirms cyberattack on email servers (Bleeping Computer) Atlassian Patches Critical Apache Tika Flaw (SecurityWeek) Microsoft: December security updates cause Message Queuing failures (Bleeping Computer) 16TB of MongoDB Database Exposes 4.3 Billion Lead Gen Records (Hackread) MI6 chief warns 'front line is everywhere' and signals intent to pressure Putin (The Record) Saviynt raises $700 million in Series B growth equity financing. (The CyberWire Business Brief) Last-minute cybersecurity and privacy gifts your friends and family won't hate (This Week In Security) Share your feedback. What do you think about CyberWire Daily? Please take a few minutes to share your thoughts with us by completing our brief listener survey. Thank you for helping us continue to improve our show.  Want to hear your company in the show? N2K CyberWire helps you reach the industry's most influential leaders and operators, while building visibility, authority, and connectivity across the cybersecurity community. Learn more at sponsor.thecyberwire.com. The CyberWire is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

Greg Coleman's career took a turn that almost no one expects. After helping build a venture-backed tech company and appearing on Shark Tank, he shifted into a world most founders overlook: government contracting. In this conversation, Greg explains how startups, consultants, and small businesses can position themselves to work with federal agencies, navigate complex programs like SBIR and OTAs, and understand what it actually takes to break into the government market.Greg spent years inside the Department of Defense innovation ecosystem, where he evaluated early-stage technologies, managed prototype programs, and worked directly with founders trying to sell to the government. Today he advises companies on how to approach the federal market, avoid common mistakes, and build real opportunities inside agencies.If you're exploring government contracting for the first time, wondering how companies get funding, or trying to understand what separates successful federal vendors from everyone else, this interview gives you a clear, realistic starting point.Chapters00:00 – Greg's background and early Air Force career02:15 – Flying high-level government officials and global missions04:05 – Launching a startup and appearing on Shark Tank07:10 – Entering the government innovation ecosystem (DIU, NSIN)13:45 – How SBIR and STTR really work for small businesses18:20 – OTAs and how companies move from prototype to production25:10 – Examples of emerging tech companies building for the government31:20 – The hardest challenge: crossing the “valley of death”35:00 – Greg's advisory work helping companies approach the federal market38:30 – Greg's thoughts on the GovClose Certification ProgramWork With GregGreg advises early-stage and growth-stage companies (Pre-Seed through Series B) on entering the federal market, building repeatable sales strategies, and navigating SBIR, OTA, and prototype pathways.Connect: https://www.linkedin.com/in/gregorycoleman/Become a Certified Government Contracting ProfessionalLearn federal sales, pipeline building, and modern acquisition strategies inside the GovClose Certification Program:https://govclose.comHire a GovClose-Trained ConsultantCompanies can get matched with trained federal sales consultants here:https://match.govclose.com

Produced by Foundersuite (for startups: www.foundersuite.com) and Fundingstack (for emerging manager VCs: www.fundingstack.com), "How I Raised It" goes behind the scenes with startup founders and investors who have raised capital. This episode is with with Naveen Verma of EnCharge AI, a startup developing energy efficient analog in-memory-computing AI chips. In addition to being CEO, Naveen is a Professor at Princeton and so we discuss his journey from academia and research to leading a startup to an over-subscribed $100 M Series B. Learn more about EnCharge at https://www.enchargeai.com/ EnCharge most recently raised over $100 million in Series B funding. The round was led by Tiger Global and included participation from Maverick Silicon, Capital TEN, SIP Global Partners, Zero Infinity Partners, CTBC VC, Vanderbilt University, Morgan Creek Digital, and others. Previous investors participating in the Series B round include RTX Ventures, Anzu Partners, Scout Ventures, AlleyCorp, ACVC, and S5V. The round also included strategic investors including Samsung Ventures, the corporate venture capital arm of Samsung, HH-CTBC, a partnership between Hon Hai Technology Group (Foxconn) and CTBC VC, In-Q-Tel (IQT), the not-for-profit strategic investor advancing technologies for the U.S. national security community and America's allies; RTX Ventures, the venture capital arm of RTX, a leading manufacturer of aerospace and defense systems and technology solutions; and Constellation Technology Ventures, the venture capital arm of Constellation, the nation's largest producer of clean, emissions-free, reliable energy. How I Raised It is produced by Foundersuite, makers of software to raise capital and manage investor relations. Foundersuite's customers have raised over $21 Billion since 2016. If you are a startup, create a free account at www.foundersuite.com. If you are a VC, venture studio or investment banker, check out our new platform, www.fundingstack.com a startup developing proprietary analog in-memory-computing AI chips

The UK Space Agency (UKSA) announces new funding to drive sovereign space innovation. The US administration's nominee for the NASA leadership role, Jared Isaacman, appeared in front of Congress today. Nuclear energy startup Antares has raised $96 million in a Series B funding round, and more. Remember to leave us a 5-star rating and review in your favorite podcast app. Be sure to follow T-Minus on LinkedIn and Instagram. Selected Reading UK Space Agency invests £17 million to drive next wave of space innovation - GOV.UK Scottish space innovation secures UK Space Agency investment - GOV.UK Trump's NASA pick to tell Congress about moon race with China, deep-space ambition- Reuters Antares Raises $96 Million in Series B Funding to Accelerate Nuclear Microreactor Development A Letter from Our CEO – Antares $96M Series B  China's LandSpace fails to complete reusable rocket test- Reuters Космонавта Артемьева исключили из экипажа Crew-12. Он фотографировал документы SpaceX и «вынес в телефоне» секретную информацию — источники Hundreds of Porsche Owners in Russia Unable to Start Cars After System Failure Share your feedback. What do you think about T-Minus Space Daily? Please take a few minutes to share your thoughts with us by completing our brief listener survey. Thank you for helping us continue to improve our show.  Want to hear your company in the show? You too can reach the most influential leaders and operators in the industry. Here's our media kit. Contact us at space@n2k.com to request more info. Want to join us for an interview? Please send your pitch to space-editor@n2k.com and include your name, affiliation, and topic proposal. T-Minus is a production of N2K Networks, your source for strategic workforce intelligence. © N2K Networks, Inc. Learn more about your ad choices. Visit megaphone.fm/adchoices

Crypto funding rounds often look glamorous from the outside: big name investors, big valuations, big narratives. But behind the scenes, the terms can look very different — and sometimes, radically so. In this episode of Bits + Bips, host Steve Ehrlich sits down with reporter Jack Kubinec, who broke the story about Berachain's Series B and one of the most unusual terms we've seen in a major token deal: a lead investor receiving the right to ask for its entire $25 million investment back, for up to a year after Berachain's token launched. Jack walks through what the documents show, why lawyers say the clause is extremely rare, and how a refund right like this could impact other investors, and even trigger MFN clauses. They also unpack Berachain's market struggles since TGE, the state of the Nova Digital fund inside Brevan Howard, and the transparency questions this episode raises across crypto venture investing. Read the full story here on Unchained Thank you to our sponsor Uniswap!  Host: Steve Ehrlich, Executive Editor at Unchained Guest: Jack Kubinec, Crypto Journalist and Podcast Host Timestamps: 0:00 — Start 0:25 — Steve introduces Jack 2:24 — What the documents reveal 5:17 — Why Brevan Howard's refund is a big problem 9:21 — How refund clauses really work 14:09 — Jack's interactions with the Bera team and how Smokey responded to the story 19:29 — Why the MFN clause is key 26:19 — How Breva Howard Digital didn't actually invest in Bera 30:18 — What investors should learn from a deal like this Learn more about your ad choices. Visit megaphone.fm/adchoices