Podcasts about Hermes

Topics covered in this episode: Web UIs for your reverse proxy Wagtail 8.0 is hot off the presses RISC-V is now officially supported by CPython Django's annual releases make every version an LTS Extras Joke Watch on YouTube About the show Sponsored by Logfire from Pydantic: pythonbytes.fm/logfire Connect with the hosts Michael: Mastodon / BlueSky / X / LinkedIn Calvin: Mastodon / BlueSky / X / LinkedIn Show: Mastodon / BlueSky / X Join us on YouTube at pythonbytes.fm/live to be part of the audience. Usually Tuesday at 7am PT. Older video versions available there too. Finally, if you want an artisanal, hand-crafted digest of every week of the show notes in email form? Add your name and email to our friends of the show list, we'll never share it. Michael #1: Web UIs for your reverse proxy Traefik, nginx, and Caddy all sit in front of a lot of self-hosted infrastructure, and all three are configured by hand-editing files. Three active projects put a control plane on top: Traefik Manager (Python + Flask), Nginx UI (Go + Vue), and caddy/ui (React + Node). All three are additive rather than replacements - none of them take ownership of your config away from you - which is the part that matters when the thing has write access to production routing. Traefik Manager is the Python one: Flask 3.1 and Gunicorn for the control plane, a lightweight Go agent for remote instances, currently v1.10.0 with an Android companion app. Nginx UI is a single Go binary at 11.3k stars, with a block-style config editor, an Ace editor doing LLM completion on nginx syntax, and an MCP server so agents can drive it. caddy/ui runs as two containers next to your existing Caddy, reads and writes your Caddyfile directly, and uses Caddy's /adapt API to validate before reload - no Docker socket required. Each one edits the config the underlying server already reads, so your files stay the source of truth and you can drop the UI without unwinding anything. Undo is a first-class feature across all three - timestamped backups with optional Git history, config version compare and restore, Caddyfile snapshots with one-click rollback. Observability is where they diverge: Traefik Manager does CrowdSec and a visual route map, Nginx UI does server metrics, caddy/ui streams access logs over SSE and pulls p50/p95/p99 off Caddy's Prometheus endpoint. Maturity spread is wide - Nginx UI has 11.3k stars, caddy/ui has 4 and was built in a single Claude session - and caddy/ui ships with auth off by default, so set CADDY_UI_USER and JWT_SECRET before it goes anywhere near a public interface. Calvin #2: Wagtail 8.0 is hot off the presses Link: https://github.com/wagtail/wagtail/releases/tag/v8.0 Custom base page models are now supported, so projects aren't locked into subclassing Wagtail's Page as shipped (Matt Westcott). New v3 REST API handles both read and write CMS operations, a first for Wagtail's API. A global registry for permission policies, plus full customizability for the remaining page views via PageViewSet. AVIF and WebP images are no longer auto-converted to PNG by default, a real behavior change to watch on upgrade. Five security fixes: page admin API restrictions, document identification by SHA1 hash, descendant collections in the Documents/Images API, snippet copy permissions, and the page translation endpoint. Formalized Django 6.1 support, and CI now runs on uv with a lockfile. Sponsor: Logfire from Pydantic Your AI agent failed at 2am. Was it the model? A tool call? The database? Most observability tools can't tell you, because they only see part of your stack. Pydantic Logfire sees all of it. One trace across your agents, LLMs, APIs, and database. Down to the infrastructure: services, Kubernetes, and hosts. It's built on OpenTelemetry, with SDKs for Python, TypeScript, and Rust, and it works with any OTel-compatible language. Every prompt, token count, and cost, right next to your vector searches and API calls. You query everything with Postgres-compatible SQL. And so can your coding agent, through the Logfire MCP server. Stop guessing. Read the trace. Pydantic Logfire. AI, it's still just engineering. Visit pythonbytes.fm/logfire today and sign up today. Get 10M records free every month, no card required. You can even click “Onboard with your coding agent” to copy a prompt to have claude or codex integrate Logfire into your app. Thanks to Pydantic for supporting the show. Calvin #3: RISC-V is now officially supported by CPython Link: https://blog.python.org/2026/08/riscv-now-officially-supported/ CPython added RISC-V as a tier 3 platform under PEP 11, specifically the 64-bit Linux target riscv64-unknown-linux-gnu. RISC-V is an open ISA anyone can implement, unlike x86 and ARM, and its market is projected to quadruple by 2032. The RISE Project donated real RISC-V machines for buildbots; the author's work was funded by a Sovereign Tech Agency fellowship. What changes: the port is now a maintained compatibility target, so CPython changes are less likely to quietly break it. What doesn't: no python.org installers, no binary wheel parity for native extensions. Next up: RISC-V runners in CPython CI for pre-merge feedback, then a push toward tier 2, plus architecture-specific optimizations. The ask is testing. If you have RISC-V hardware, build CPython, run your test suite, file what breaks. Tier 3 is the weakest support tier. PEP 11 tier 3 requires a core developer contact and a buildbot, but failures on tier 3 platforms explicitly do not block a release. Saying "ongoing CI/testing expectations" oversells it. The honest bit is "someone is now on the hook for it, and breakage gets noticed," not "it's guaranteed working." Worth the caveat that this is Linux SBCs, not microcontrollers. A VisionFive 2 counts, an ESP32-C6 or Pico 2 does not. Those are 32-bit non-Linux parts where MicroPython is still the answer. Michael #4: Django's annual releases make every version an LTS Starting with Django 2028, Django will move to one January feature release per year, adopt calendar-based version numbers, and support every release for three years. The old distinction between standard and LTS releases disappears, giving teams a predictable annual upgrade path that aligns more closely with Python's own release and support cadence. Every Django release becomes the safe, long-supported choice, so teams no longer need to wait for a specially designated LTS version or absorb two years of changes at once. Each release gets one year of mainstream bug fixes followed by two years of security and data-loss fixes. New releases support the three latest Python versions and add the next Python release during their first year. Calendar versioning begins with Django 2028, followed by Django 2029 and so on. Three Django versions will be supported at any time, giving third-party packages a clearer rolling target. Nothing changes before 2028, and existing commitments for Django 5.2 LTS and 6.2 LTS remain in place. Extras Calvin: The Python docs now document the time complexity of built-in types https://docs.python.org/3.16/library/time-complexity.html Thinking in Python - Bruce Eckel's free book https://thinkinginpython.com/ Michael: prune_uv_pythons.py - Prune uv-managed Python installs, keeping only the newest patch per minor version Runs automatically in my system “upgrade” script: upgrade-output-2026.png Started using Ollama cloud models for my Hermes assistant. Thanks to Jeff Triplett I learned they are not just local models. Joke: The Tao of Programming - Book Seven: Corporate Wisdom

Send us some Fan Mail? Yes please!They're back! Was it the Alien and the Amputee, or the Amputee and the Alien? Who even cares anymore? At least you know it'll be anything but sane or safe. Enjoy. Connect with Nick: His podcast, YouTube, & Twitter. Connect with Hermes: Instagram & Twitter Subscribe, rate us 5, come join in all the other fun we offer, but most of all we hope you enjoy! If you liked this, and want to hear more, give us a follow and let us know! Or maybe you just want to tell us how awful we are? Comments help the algorithm, and we love to see ‘em! And as always, don't kill the messenger. Whiskey Fund (help support our podcast habit!): PayPalOur Patreon & YouTube Support the show

Eine sehr politische Folge – und eine, in der ich mich für Max Monolog-Lastigkeit der letzten Wochen revanchiere.Zum Berliner Landesnetz: Die Systeme sind wieder online, Bürgerservices erreichbar. Digitalstaatssekretär Florian Hauer sagt, es gebe aktuell keinen Hinweis auf eine Infiltrierung, aber nur als Momentaufnahme. Abgeflossen seien nur ohnehin öffentlich verfügbare Daten. Wir diskutieren, ob diese Art halbgarer Kommunikation hilft – und ob man es in so einer Lage überhaupt richtig machen kann.Das Hauptthema: Am 12. August hat Trump ein National Security Presidential Memorandum unterzeichnet, das geprüften US-Unternehmen offensive Cyberoperationen gegen ausländische cyberkriminelle Organisationen erlaubt – unter staatlicher Kontrolle, mit Verträgen bei DOJ oder DHS und schriftlicher Freigabe pro Operation. Ich bin überrascht, wie durchdacht das Papier formuliert ist, gerade bei den Definitionen. Nur ist damit noch nicht mal die halbe Miete gemacht: Die eigentlichen Verfahren, also Mindeststandards, Targeting, Deconfliction und Rules of Engagement, müssen erst in den nächsten 60 Tagen definiert werden, und bis dahin darf keine einzige Operation genehmigt werden. Offen bleiben bis dahin Attribution, Third-Party-Infrastruktur, Haftung bei Kollateralschäden und ein dünnes Oversight-Modell ohne Berichtspflicht an den Kongress. Meine Vermutung zu den Teilnehmern: Die Großen übernehmen risikoarme Botnet- und Scam-Compound-Takedowns, für alles Heiklere entstehen Startups.Max bringt eine Keycloak-Schwachstelle: CVE-2026-18963, CVSS 9.1, unauthentifizierte Account-Übernahme über einen schwachen Password-Reset-Mechanismus. Alle Versionen bis 26.7.2 betroffen.Zum Abschluss zwei Meldungen mit möglichem Staatsbezug: Ein kleiner britischer Stromerzeuger war im Juli vier Tage offline, Medienberichte deuten auf iranische Akteure, offiziell attribuiert ist nichts. Und Taiwan: Die israelische Firma Dream hat ein 160-MB-Archiv analysiert, das eine viertägige Kampagne mit bis zu acht parallelen Subagenten auf Basis von OpenClaw und Hermes dokumentiert. 21 Regierungssysteme kartiert, 85 Accounts geknackt, betroffen auch die Nuklearsicherheitsbehörde. Der Initial Access war banal: drei vergessene Debug-API-Endpunkte. Faszinierend ist die Orchestrierung – Findings wurden mit Wahrscheinlichkeiten bewertet, bei Sackgassen recherchierten die Agenten selbstständig neue Techniken.NSPM "Expanding Capabilities to Combat Transnational Cyber-Enabled Crime" (Weißes Haus) https://www.whitehouse.gov/presidential-actions/2026/08/expanding-capabilities-to-combat-transnational-cyber-enabled-crime/Juristische Einordnung des Memorandums (Mayer Brown) https://www.mayerbrown.com/en/insights/publications/2026/08/presidential-memorandum-authorizes-vetted-private-companies-to-conduct-offensive-cyber-operations-against-foreign-criminal-organizationsOffene Fragen zu Haftung und Oversight (Crowell & Moring) https://www.crowell.com/en/insights/client-alerts/license-to-hack-the-white-house-greenlights-private-sector-offensive-cyber-operationsKeycloak CVE-2026-18963 (Red Hat) https://access.redhat.com/security/cve/CVE-2026-18963Dream: "Inside a Multi-Agent AI Framework Used to Compromise Government Entities in Asia" https://www.dream.security/blog/inside-a-multi-agent-ai-framework-used-to-compromise-government-entities-in-asiaTaiwan Ministry of Digital Affairs: Monatsbericht Cybersicherheit https://moda.gov.tw/en/press/monthly-report/Einordnung des Taiwan-Angriffs (The Register) https://www.theregister.com/security/2026/08/12/near-autonomous-ai-agents-attack-taiwans-nuclear-safety-agency/Berliner Landesnetz: Update der Senatskanzlei https://www.berlin.de/rbmskzl/aktuelles/pressemitteilungen/2026/pressemitteilung.1703898.phpOpenAI: "Pacing model development in an era of cyber-critical capabilities"https://openai.com/index/pacing-model-development-cyber-capabilities/

Hey this is Alex, welcome to... the chillest week in AI, since ... a long time. Chill, if you consider Moderna and MERK announcing a cancer vaccine and surging 115% in a day, a chill week. This week, the only two model drops we really saw came from the excellent Z.ai folks, they announced GLM 5.3, API only for now, and an amazing tiny release of Qwen 3.89 27B. In other big AI news, OpenAI announced they are pausing RL efforts (Reinforcement Learning) to focus on security and alignment post the scary AI Swarms hacking incident, dedicating up to 20% of compute towards reviewing agent thinking processes, and Stripe buying OpenRouter for a reported $8B! Sometimes the chill weeks are actually good, we're able to chat about how we use AI, what changed for us, and give our guests a bit of breathing room. This week, I invited Francesco from CUA to talk about computer use in open source + their new history plugin, Bin from HeyGen to talk about HyperFrames, a way for your agents to create videos and a breaking news guest, Jeff Huber from Chroma jumped on to talk about their new Foundations release, a unified memory for your agents! This was a great episode, I hope you'll like it, it's up here on Substack and everywhere you get your pod (Spotify, Youtube, Apple Podcasts). ThursdAI - Highest signal weekly AI news show is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.Are we being fed slop again? (Is Claude dumb again?)Before we get to releases, this week on the show, I complained, again, that I feel my AI's are degrading. If this feels like de-ja-vu to you, it's because the same happened a year ago in September 2025 (and Anthropic admitting this 2 weeks later), and ... now this happens with Fable?You see, I use pretty much the same prompts, every week, preparing for the show. This is partly my way to evaluate new models and compare to existing and previous ones while also bringing you the best researched weekly show in AI. Well, this week, one after another, Claude Fable, which is... like the best intelligence, gave me such poor output, that I couldn't believe what I'm seeing. First, literally ignoring instructions that say “hey, show me all the items I've collected and let me pick the most important ones”, Fable instead sent all of them to my research pipeline, without showing me. This has worked, consistently, without fail, for the past... year? maybe more! This worked with open source models, worked with GPT, and now Fable, a Mythos Level LLM, is doing the most basic dumb s**t possible, ignoring the main reason I even have this workflow. And this wasn't just a fluke either, when asked to create a run of show document, and given an example, Fable produced this... whatever this is. This is the same document and same format that Fable produced for me during AI Engineer which got me thinking “ok, this is AGI”, and here, given an example, I got a completely unusable artifact, despite direct instructions, structure and example! I got to say, given that privately this week, Anthropic disclosed that they have passed $65B in revenue, which is absolutely insane, this doesn't add up. So I figured, ok Alex, maybe this is your prompts or skills. But no, LDJ came in with some charts that show degradation, one from MarginLab.ai that shows significant lowering on number of tool calls and average runtime recently (this is for Opus 5) and And another chart from modelverify.ai model drift monitor showing drift scores.Do we have anoher Claude Gate on our hands? Is your Fable/Opus behaving weird lately? Or did you completely switched away to other models? OpenAI pausing RL and focusing on safetyLook, when we covered the HF hacking incident and then the pacing the frontier letter, I didn't imagine that results will come this fast, but this week, OpenAI publicly announced that they are pausing RL training, which is the last step of models, until they get their sandboxes in order and align the models better. We all agreed on stage that this is likely a very good move, and Peter was really awe-struck at the 20% dedication of resources towards reviewing thought processes of models. Is this a good enough response to the scary hacking incident? we'll see, but I think this is the right move from OpenAI, and still, waiting for the full postmortem on the OpenAI security incident. Open Source LLMsQwen3.8-27B ties GPT-5.6 Luna and runs on a 4090 (X, HF, Announcement)Following the release of their flagship, Alibaba dropped a model that became a community darling overnight, Qwen 3.8 with just 27B parameters. This “tiny” model scores 52 on the Artificial Analysis Intelligence Index, same score as GPT 5.6 Luna at Max reasoning and 51 on Agentic index, beating Opus 4.8 MaxAll while running at around 68t/s on a 4090 GPU, and around 40 on max via MLX, hell it even does 11t/s on Xenova's WebGPU kernels right in the browser! This model exploded on the HuggingFace hub, with tons of quants, over 152 fine-tunes, it was downloaded over 10M times overall

I stick my neck out on this Black Iron Prison Intercept, or do I? I'll take you on a journey that uncovers the Craft as a lived Hermes cult, revealing an esoteric Gnosis where the Lodge floor serves as a ritual cartography of the soul's entrapment within the material prison of the Demiurge. By receiving the Light from the East, the initiate awakens a divine spark required to transcend this material lie and return to the true, transcendent Deity. These spiritual mysteries are woven into the Cosmic Matrix of star lore, where the precession of the equinoxes and planetary elevations rhythmically dictate the rise and fall of global religious eras. Finally, the high degrees bridge the gap between antiquity and the Middle Ages, identifying Freemasonry as the mystical continuation of the Knights Templar through a chronological synchronization known as the Mirror of Time. Get The Occult Elvis: https://amzn.to/4jnTjE4 Virtual Alexandria Academy: https://thegodabovegod.com/virtual-alexandria-academy/ Gnostic Tarot Readings: https://thegodabovegod.com/gnostic-tarot-reading/ The Gnostic Tarot: https://www.makeplayingcards.com/sell/synkrasis Homepage: https://thegodabovegod.com/ Patreon: https://www.patreon.com/aeonbyte AB Prime: https://thegodabovegod.com/members/subscription-levels/ Voice Over services: https://thegodabovegod.com/voice-talent/ Support with donation: https://buy.stripe.com/00g16Q8RK8D93mw288 Merch store: https://aeonbyte.creator-spring.com/ Equipment Wishlist: https://www.amazon.com/hz/wishlist/ls/2WEJ2CCWHALZB?&sort=default   Intro concept, visual direction, and AI-assisted creative development in collaboration with Arturo Pérez E. youtube.com/@333amTV 333am.tv Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Send us some Fan Mail? Yes please!We know that none of you asked for it, but you're unfortunately stuck suffering alongside solo Captain Hermes this week. Seems life has been a little bit of a bully as of late, but thankfully we're no strangers to a few sadness sprinkles garnishing this show dear listeners. The sun always rises eventually, shows the way out of the valley, if only just in the nick of time. Until next time friends, cheers. Subscribe, rate us 5, come join in all the other fun we offer, but most of all we hope you enjoy! If you liked this, and want to hear more, give us a follow and let us know! Or maybe you just want to tell us how awful we are? Comments help the algorithm, and we love to see ‘em! And as always, don't kill the messenger. Whiskey Fund (help support our podcast habit!): PayPalOur Patreon & YouTube Connect with Hermes: Instagram & Twitter Support the show

Amazon keeps raising the bar. Steve Dennis and Michael LeBlanc open with the retailer's blockbuster quarter: massive gains in AWS and advertising, staggering AI capex, and retail revenue up 16% year over year, up from 12% last year. Third-party sellers now drive more than 60% of the business, and Citi pegs underlying GMV growth near 10%, roughly double the industry average. Amazon's B2B division alone now tops $60 billion, bigger than all but eleven U.S. retailers. Then the hosts dig into the agentic commerce numbers Amazon disclosed as Rufus folds into Alexa shopping, why sample bias demands caution here, and how grocery momentum is turning Amazon into a mega market-share threat.  Then: Shopify's revenue up 32%, and the tale-of-two-cities reality that a handful of giants now drive nearly all e-commerce growth. A mid-year check on Steve's Prediction #8 — luxury's future won't be evenly distributed. LVMH, Kering, and Capri lag on China softness and war induced Gulf weakness, while Hermès, Ralph Lauren, Richemont, and Zegna keep delivering outsized results. Live from the CommerceNext Growth Show: Kartik Hosanagar. He's Wharton's John C. Hower Professor of Technology and Digital Business, author of A Human's Guide to Machine Intelligence, and co-founder of Bliss Labs. His argument: the biggest shift in retail isn't a new technology or channel. It's a new customer. Treating AI as another distribution channel is the same mistake movie studios made with Netflix. There's no marketing science for AI yet. $9.99 pricing, scarcity, social proof — a model may not respond to any of it. Kartik walks through the Ridge wallet case: invisible in ChatGPT's "gifts for men" results, then the default recommendation within three weeks, complete with a hallucinated promotion. Every retailer faces the same fork: efficiency or meaning. The middle is the most dangerous place to stand. Kartik also details the simulation sandboxes that let brands test counterfactuals in a day instead of eight weeks, why you can't just ask an LLM why it picked a brand, and the three attitudes retailers need now: curious, experimental, collaborative Back in studio: Wayfair's encouraging quarter and Steve's change of heart, Warby Parker's mixed report as store count passes 400, tariff rebates ($100 billion of $160 billion already repaid, more tariffs looming), and whether live selling has hit its tipping point with QVC out of bankruptcy and Whatnot valued at $20 billion. About UsSteve Dennis is a strategic advisor and keynote speaker focused on growth and innovation, who has also been named one of the world's top retail influencers. He is the bestselling author of two books: Leaders Leap: Transforming Your Company at the Speed of Disruption and Remarkable Retail: How To Win & Keep Customers in the Age of Disruption. Steve regularly shares his insights in his role as a Forbes senior retail contributor and on social media.Michael LeBlanc is a senior retail advisor, keynote speaker and media entrepreneur. Michael has delivered keynotes, hosted fire-side discussions hosted senior retail executive on-stage in 1:1 interviews worldwide. Michael produces and hosts a network of leading retail trade podcasts, including The Remarkable Retail Podcast, The Voice of Retail, The Food Professor, The FEED powered by Loblaw and the Global eCommerce Leaders podcast. He has been recognized by the NRF as a global Top Retail Voice for 2025 and 2026 and continues to be a ReThink Retail Top Retail Expert for the fifth year in a row.

An old Linux box powers down for the last time as we unwind its history, and the entire show, back to the beginning.Sponsored By:Jupiter Party Annual Membership: Put your support on automatic with our annual plan, and get one month of membership for free!Managed Nebula: Meet Managed Nebula from Defined Networking. A decentralized VPN built on the open-source Nebula platform that we love.Support LINUX UnpluggedLinks:Web Boost — Send us a boost via sats or USD

https://youtu.be/eUqMjIq3r2U Brian Hong, Founder of Infintech Designs, Flowbots.ai, and BigEasyData.ai, is driven to Build AI Revenue Engines that create freedom by helping businesses generate more leads, sales, and revenue while increasing the productivity of their teams. By combining SEO, AI automation, business intelligence, and human expertise, Brian helps companies improve their marketing, systemize their operations, and enhance their people with AI rather than simply replacing them. In this conversation, Brian introduces his Team Building Framework—Attract and Assess A-Players, Productize and Systemize, Hire People Who Can Follow Instructions, and Enhance with AI Capabilities. He explains how placing people in roles where they can thrive creates stronger teams, why clear processes and blueprints make it easier to scale, and how AI can enhance employees by increasing their productivity instead of replacing them. Brian also discusses how SEO is evolving with ChatGPT and Google AI Overviews, why businesses need to build brand authority and entity breadth, and how business intelligence can identify the specific areas where AI can create the greatest revenue impact. — Build AI Revenue Engines with Brian Hong  Good day, dear listeners. Steve Preda here with the Management Blueprint Podcast, and today my guest is Brian Hong, the founder of Infintech Designs, Flowbots.ai, and BigEasyData.ai. These are New Orleans-based companies, including a digital marketing agency, focused on building more leads, more sales, and more exposure online with strategic SEO and digital marketing strategies. Brian, welcome to the show.  Thanks. Thanks for having me.  So you have quite a portfolio. I also read on your LinkedIn page that you invest in other companies, and then you support them as well. So it sounds like you have a lot of irons in the fire. But before we get into all that, my question to you is, what is your personal “Why,” and how are you manifesting it in your business, or businesses?  Why do I work so hard? Why do I want to build all these companies? To me, money creates freedom. I’m trying to bank as many freedom points as I can to live life on my own terms, and to do and have experiences with the people I love. That’s my “Why.”  Wow. How does that manifest in your businesses?  That’s a good question. I'm building AI automation to get my time leverage, to create duplication, so I can 10x the productivity of my employees. Not hire more people, but keep my A-players.Share on X Because if I have an A-team, or can cultivate an A-team, then I think I can bank more freedom points at a more efficient and scalable process. Don’t hire more people. Hire people intentionally, and 10x their productivity. That would be a domino effect to align with gaining more freedom points.  So what’s the maximum number of freedom points? Is it like a 100-point scale? What does it look like to have 100 freedom points?  It’s working because I want to, not because I have to. Because I love working. It’s living life on my own terms and focusing my time on things that I want to do. For instance, I have a house cleaner. I don’t want to clean the house. My time is better spent doing something else that I passionately want or love to do. Unless I loved cleaning houses. That’s fine. Maybe that’s therapeutic for some people. But it allows me to live life on my own terms. Money is the key that unlocks those opportunities to live life on my terms.Share on X  How do you maximize your freedom points? What does it look like when your freedom points are maxed out?  Then I’m working because I want to, not because I have to. I still want to work. I still want to build. I love building things. I’m not the type of guy who wants to go to the beach, sit on the beach, and do nothing all day. I might do that for an hour. I’ve got to be doing something. I have to be building. It’s just the way I’m programmed. Maybe I don’t want to work as much as I do now, but I can’t imagine a life doing nothing. To me, that doesn’t serve a purpose.  Yeah.  I like building. I love contributing. I love being a part of the equation of success. It's just fun to me. I get gratification out of it. I like creating jobs.Share on X I like creating communities. I like meeting people. It’s interesting to me. It fills my bucket. That bucket is filled to then, I mean, try and give me freedom points to then go spend some personal time creating experiences with the people I love, mainly my wife and my child.  That’s amazing. Okay, let me take a step back. You have multiple companies. You have Infintech Designs. You have a couple of AI companies as well. How does that portfolio work?  I essentially do this. I have a construction company as well, TurnKey Renovators, a GC for residential. I have a couple of e-commerce companies. Really, I’m doing the same thing over and over. Earlier in my life, I used to wait tables. My takeaway from that experience, from working at Applebee’s to fine dining, is: How many different dishes can you make with the same ingredients?  With my knowledge, my resources, and my skills, how many different things can I do with them? This is my version of it. All those companies, by the way, I’ve never stroked a check. I pretty much say, “Give me half your company, and I’ll help you grow it.” I’ve never cut a check because I give them what they really want, which is the outcome they want to achieve with that money. They want the money to help grow their company. What if I just help you grow your company?  So I ask for an equal seat at the table, and in exchange, I’ll install what I know and go grow it. I’m essentially doing that over and over. My time is spent on high-level decisions, systems, processes, automation, and marketing.  Yeah. Love it. You know, I always wondered about marketing agencies. If they really know how to grow sales, why are they not doing it for themselves, and why are they offering it to others? It sounds like you are drinking your own Kool-Aid, and you are using your own product.  Yeah. Definitely. We practice what we preach. I see some companies doing that because there’s also an abundance mindset. If I can do it for me, and I can do it for someone else, let me just create another revenue stream and do it for both of us.  Okay. That makes sense. This podcast is all about frameworks. I wonder, what is a framework that you have created or discovered that helps you be effective in what you do, and that you could share with the audience so they might also find a way to leverage it in their business?  I guess it depends on many things. From team building, it may be doing a DiSC Assessment and Kolbe tests to make sure we place each person in a situation where they thrive. Sometimes when things aren’t working, maybe the owner, the manager, or the person needs to look in the mirror and say, “This is my fault, not the employee’s fault.” “I didn’t set them up for success.” “I’m trying to stick a round peg into a square hole.” “I’m trying to put them in a position that they may not thrive in, and they’re not programmed to naturally do.”  So we’ll sometimes do these assessments, whether it’s a personality clash or a personality type, and make sure we place them in a role where they’ll thrive. So that could be a framework. Setting the employee up for success. Once we do have an employee, the better the plan, the better the outcome. So try to productize and systemize everything we do. If we have a blueprint people need to follow, then I can spend more of my time not finding specific talent and skills, but finding people who can follow instructions.  Because if we have the blueprint, and this is what you need to do, then I can get a lower-wage worker who doesn’t have a specialized skill. That specialized skill will be developed because I’m handing you the blueprint of what you need to execute.  So that’s the three-step framework. Set employees up for success by delegating tasks that fit their personality. Then productize and systemize. Then find people who can implement. To what extent does that work in the AI age? Are these people who just execute tasks according to instructions still solving the problems, or is that no longer working because of the fast pace of change?  It depends on what the activity is, but generally, and more often, it’s human enhancement, not human replacement. So instead of saying, “Let’s just say I have a web developer, and I get more business, so I need to hire a second web developer, web designer, programmer, then a third one, then a fourth one.” The new version today is: keep the one you have, and instead of hiring another human, 10x their productivity with an AI-driven solution.  Allow them to do more with less. Allow them to duplicate their activities so they focus on their revenue-generating activities, on the decisions and activities that AI cannot do, as of yet. That’s how we turn an A-player into an A-plus player. That’s how we turn a B-player into an A-player. That prevents me from having to hire people I don’t want working for me, which are C-players.  Isn’t there a risk that when you have too few players, even if they are A-plus, there’s too much dependence on a single person? Well, I mean, we’ll hire more than one. I have a seven-person leadership team. I have two main managers. I have a department head and then an associate, so there’s always some level of duplication. Then you have an AI layer in between that. So between all of that, yes, if someone gets struck by lightning or someone is sick, the ship needs to keep sailing. We can’t shut down the business.  Love it. What is driving growth in your business?  Relationships. SEO. Ranking in LLMs. Google AI Overviews. ChatGPT. I’d say those are the main pillars that are creating opportunities. I also own a conference, rockstarsconference.com. We’re entering our 15th year. That’s network value, positioning of expertise, and relationship building. Things that AI cannot replicate. Human connections. Then I also do SEO for myself, right? Our last handful of leads that came in through Flowbot.ai. This one particularly stands out. There was this roofing company that sent me an email. We actually talked on the phone.  He really blew up my ego and talked about how great I was. I thought he was a referral. I’m like, “Man, why does this person think I’m so great? It’s really odd.” I said, “Can I ask how you heard about me?” He’s like, “Oh, I spent the last two hours talking to Gemini because I discovered you through Google AI Overviews, and it just told me you’re great.” So I’m like, we’re entering a new world. That is a referral. It’s not from someone I know who spoke highly about me. It’s through a language model. Gemini said I was great, so this person believed I was great.  That’s amazing. That’s very cool. So you have these four pillars. I love it. Relationships, SEO, GEO—or whatever AI referrals you want to call it—and the conference. Would you bucket the conference into the same category as relationships, or is that more of a proactive way of creating a community beyond personal relationships? Is it the same thing?  It checks a lot of boxes. A few of my business partners were born out of that conference. New opportunities and getting clients were born out of that. Knowledge gain. Knowledge transfer that I gained to go implement. So it checks all the boxes. I’m gaining knowledge. I’m gaining relationships. I’m gaining partners. I’m gaining new systems. New processes. I’m gaining sparks, catalysts, and kind of tracks to go down and explore. So it kind of checks a lot of boxes.  Yeah. Love it. So the SEO and GEO piece, is it something that is static, or is it dynamically changing all the time?  Dynamically changing. It’s SEO 3.0, right? SEO is dead. It’s dead if you’re doing SEO the old way. Getting ranked in ChatGPT, the LLMs, and Google AI Overviews. Let’s just say it’s everything you do with SEO, plus more now because you have ChatGPT citations. You have sources that these LLMs frequently query. Then, in the world of SEO—the search engine Google—you have broad core updates. Broad core updates. Updates to the algorithm. That’s pretty much what it means.  They also occur within LLMs. So it’s pretty much do SEO, plus more. What is the plus more? If I had to distill it, it’s get brand mentions and actually become a brand. Do podcast interviews like this. Get press releases. Get write-ups. Be on SoundCloud. Do social media. Post on LinkedIn. Get write-ups in news and media publications. Appear in Google News. Have a website that has dynamic content. Create your topical map. The list keeps on going. Build your brand. That’s the SEO of today. That will put you in a position to get mentioned in ChatGPT and Google AI Overviews.  Yeah. That’s fascinating. I’m just thinking about my business. We had a website that actually got some referrals from AI. We even got some clients last year. Then we split the website into two websites, and things dried up. It was kind of an interesting situation. We still had the same amount of content, but we moved it to a different domain, and it did not register the same way. Do you see that often?  Yeah. You want Google and LLMs, at the most basic level, to have a better understanding of who you are, what you do, where you do it, and the services you offer through having something called breadth of entities. Entities are the nouns of the internet. Entities are a signal. A language. All Wikipedia pages are entities, but not all entities have Wikipedia pages.  They’re the nouns of the internet. It’s a layer of clarity for search engines, algorithms, and LLMs to understand who you are, what you do, and where you do it. It’s not the only signal, but at a foundational layer, that should be implemented. So what we’re talking about is creating content. What does that mean? Create content on your website. Don’t create content for the sake of creating content because anybody can create content with AI.  Create intentional content that satisfies the user and aligns with certain algorithmic components, such as Google's E-E-A-T: Experience, Expertise, Authoritativeness, and Trust.Share on X Everything we’re talking about is SEO. You want optimized content. You want pages. The first 150 to 200 words are important. The last 150 to 200 words are important. Add tables. Add bullet points. Make sure your content is clear. Position yourself with authority.  Have links pointing to it. Interlink it. These are all SEO kinds of activities. So SEO is alive and well, but there are new layers you need to add on top of that to create optimal positioning in Google AI Overviews and ChatGPT.  Yeah. I saw one of your videos on your LinkedIn page where you actually explained what is the wrong way to do SEO and the right way to do it. I think you had your alter ego disputing with you. Yeah. That was pretty funny. So that sounds like a lot of work. Can you delegate that to AI agents to do this kind of stuff?  Yes and no. It’s not like I just click a button and step away. It takes a human-in-the-loop system, and that’s what we’ve developed. So yes, I don’t need as many humans, but I still need humans.  So you said that in one of these companies you have 100 AI agents working on SEO, creating SEO impact for your clients. What is it that the AI agents can do, and what is it that you still need the human to do?  The AIs collect information, rapidly digest it, find patterns, perform pattern analysis, do research, and create content briefs. But there needs to be thought leadership. AI is very directional. It’s, “Go do this,” but we need a human to point it in the right direction. We need to create guardrails. We need to create a source of truth. It needs to know where to look for the right information. It needs direction on what to do with this information.  How should I digest it? How should I format it? How should I break up the heading tags? It needs that element where it can go do it, but the human is part of that last-mile conversion. So maybe the AI does zero to 80, and maybe the human does the 80 to 100.  Yeah. That’s very interesting. So what’s one thing that you’re actively trying to figure out in your business?  How to automate everything. Everything is a complete agentic workflow that is proactive, not reactive. Where I have a board of advisors every day looking at how they can serve me.Share on X Identifying the patterns in my life that I need to stop doing and they can do. Developing skills, if they don’t have them, to go execute that. Having a feedback loop to constantly improve.  This has become superintelligence. Agentic work, which is kind of here. That’s what maybe some people have heard about: OpenClaw. Then you have something called a Hermes agent. These things are here, but they need to be trained. They need to be fed the right information and given the right guardrails. But it has begun, and I’m trying to go down that journey.  Isn’t that overwhelming to think of all these details, managing these agents, and making sure they’re not going rogue or getting confused by what every one of them is doing?  It’s overwhelmingly exciting, but not overwhelmingly bad, because this is the worst it’s going to be, and it’s mind-blowingly good. It’s overwhelming because it’s so amazing and so awesome. It’s garbage in, garbage out. I am the architect trying to create the blueprint on how to control these agents. To me, that’s incredibly exciting and incredibly amazing. To be a part of this time in history, it’s like that feeling when I dropped out of college and this thing called the internet was born in 2000.  I knew the world wasn’t going to be the same again. That same feeling came again in 2022 when I heard about ChatGPT. I created my first AI company in 2023. I said the world isn’t going to be the same again. I have the same feeling I had in 2000, but times 100. So to me, it’s overwhelming, but overwhelmingly exciting.  Yeah. So what does one do if they want to get, okay, let’s say it’s me. I use ChatGPT actively, but I’m not building AI agents. What would you recommend that I do? How do I even get started?  What is your pain point? Is it communication? Is it writing emails? Is it reporting? It kind of depends, right? You can customize it based on your pain points. Where are you spending your time, and where should you be spending your time?  Yeah. That’s a good question. I think the first question is, what should I even use it for?  I can give you some ideas. A handful of things I use it for, but not limited to. I can give you use cases that we do for other people. So to construction company, inbound and outbound calls. We never miss a phone call. We answer every single one. We qualify the lead. We seamlessly connect to a CRM. We know if it’s a new customer or an existing customer. If it’s an existing customer, we can look at their history. We can now have an intentional conversation through voice.  I can clone that. I can qualify them. I can book the appointment. I can put it on the salesperson’s calendar. I can receive a text message. I can read the message. I can set the salesperson up for success. I can then pull their address. I can then pull their demographics. I know their net income. I know their affinities, their interests, their hobbies. I know whether their house has a pool. I know the number of bedrooms and bathrooms. I know their credit score. I know their disposable income.  My layer of intention is going to be very different if I have all of this data to interact with that person, either through a voice agent, an AI agent through SMS, through email, through voice, or I can pass that information to the salesperson so they know what they’re walking into. They know the opportunity. So we have lead scoring. We have prioritization. That's one thing we could do: taking over your appointment setting and lead qualification.Share on X Then we can create reporting.  Then we can create trend forecasting. We can turn your business into a mathematical equation. We know 10 leads equals 3 appointments, equals 2 shows, equals 1 person who shows up, average ticket $10,000. We know that formula. Then we say, “Well, we need to feed it more.” The math equation is 10 equals $10,000. Whatever that math equation is, that’s the position I want to be in. I want to make data-driven decisions. So AI is an accelerator to achieve that.  What’s at the top of the funnel? Is it advertising?  Yeah. It could be Facebook Ads. Google Ads. SEO. It kind of depends. Do you want expensive and fast, or do you want cheap and slow? Expensive and fast. When I say expensive, I mean cost per acquisition and cost per visibility. That’s going to be Google Ads, Facebook Ads, LinkedIn, TikTok—anything where you’re buying media. That’s going to be expensive because you have to pay on a per-click basis or a per-impression basis. The cheap and slow is going to be something like SEO.  Do the work now to feel the impact later. It’s signal building. You don’t have to pay on a per-click basis. You don’t have to pay on a per-impression basis. But you have to build signals. Those signals cost money. You have to do the work, but it’s not going to happen overnight. You do the work now to feel it later. It’s an investment. So ideally, you do both. You take a blended cost per acquisition because your SEO channels almost 99% of the time are going to have the lowest cost per acquisition. But speed is going to come from Facebook Ads and Google Ads. So you take the best of both worlds with a blended cost per acquisition.  Is the formula the same for business-to-consumer and business-to-business companies, or is it different?  Yeah. The algorithm doesn’t change. The algorithm is the algorithm. The messaging is different, but it’s the same across the board. The concepts are the same. There are some things, like in medical. Google specifically has something called YMYL—Your Money or Your Life. That’s going to be finance businesses, payday loans, and medical. Those things have a high impact on your life.  So they may have a little more scrutiny, and it might be harder to build trust and validation because of the category you’re in. In that aspect, yeah, the formula is a little bit different. But the activities you engage in are the same. You need to build content. You need to build your brand. You need to build links. You need to optimize your website. You need to create on social media. It’s all the same.  Yeah. Fascinating. So who is the ideal client that you would like to respond to this podcast?  Man, I get asked that, and I’m having more difficulty because I have a lot of companies in my portfolio, right? I have, as of now, about seven companies. I’d say companies that want to automate. Why do you want to automate? The first step of that is maybe building intelligence—business intelligence. You can’t improve what you can’t measure. What if we build a system where maybe you take two steps back to take 10 steps forward?  The two steps back is building business intelligence through AI-driven systems to have segmentation and attribution on exactly what’s happening. How many leads are you getting? Where are they coming from? What is your workflow? Create the math equation. Now we have a measurement. We have a benchmark. Now we can build custom AI agents to say, “This is what we need to focus on first.” So we have our foundational layer, our source of truth.  Maybe that’s a good first step, so you’re not doing a spray and pray. You can say, “Now I know what I need to do. Now I know where my problems are.” Let’s remove the emotion of, “I think.” It now becomes, “I know this is a problem.” Now let’s zoom in and say, “Well, how do we solve this problem?” Are leads the problem? Or is it connecting to the leads? Is it showing up to the appointment? Is it closing the deal? Are we closing deals, but the average ticket is low?  If we close one deal, is it upselling to a second deal? What is the one-year value? Lifetime value? First transaction value? If we don’t know that, then how do we create a math equation of success? How do we know what bucket we need to feed? How do we know where the problem is or isn’t? If we do know where it is, then we can say, “Let’s build a second AI agent.” How do we augment and enhance maybe lead-to-connection rate? Let’s build an SMS agent.  A voice agent. An outbound agent. Let’s build nurture and upsell. Let’s follow up in perpetuity, forever. Let’s tackle that first because we have clear, data-driven insights showing this is a problem. I would say people who want clarity on where their problem is, or maybe they know where their problems are. Then that could be a starting point. So it’s hard for me to answer because I’ve got my hands in so many different buckets.  I’m top of funnel. I’m middle of funnel. I’m bottom of the funnel. I’m top of funnel. Introduction. Brand. Product or service. Google Ads. Facebook Ads. SEO. GEO. I’m middle of funnel. Turning visitors into customers. That’s follow-up AI systems. That’s operational efficiencies. That’s touchpoints. That’s nurturing. I’m bottom of funnel. I’m business intelligence. Segmentation. Attribution. I can help with all stages. I just need to talk to businesses. Businesses that want to grow and generate more revenue. That’s my audience.  Love it. Love it. So if you have a business that wants to grow and generate more revenue, then make sure you check out Brian Hong on LinkedIn. Where else should people go if they want to learn about your companies?  Yeah. I’ve got Brian Hong Digital on Instagram. I’m trying to build my own personal brand. I’m about to relaunch some videos again. I’ve been busy with another acquisition of another marketing agency. I’d say check out the socials or send me an email through Flowbots.ai or InfinTech Design.  Okay. Sounds good. So check out Brian. I mean, that’s pretty amazing. The complexity that you’re managing. Not just the number of companies. The number of investee companies you’re in. The products you’re building. The conferences. It’s kind of mind-boggling how you’re managing that complexity. But I guess you have a couple hundred AI agents at your service. If you’d like to learn more and explore Brian’s products, go to his LinkedIn page.  Then you can connect to his different companies from there. If you enjoyed this conversation, make sure you stay tuned because every week I bring a couple of amazing entrepreneurs like Brian who will open your eyes to the opportunities ahead of you. So thanks, Brian, for sharing your knowledge and insights. Thanks for listening. Thanks, Steve. Appreciate it. Important Links: Brian's LinkedIn Brian's  website

FIRST 54:00: Spoiler-Free Review with our See It/Skip It Verdict AFTER 54:00: Full Spoiler Review It's another edition of See It or Skip It and this time Ian is joined by JDG, Juleen and Hermes as they swing headfirst into Spider-Man: Brand New Day — Tom Holland's latest outing as Peter Parker and the beginning of a very different chapter for the MCU's resident wall-crawler. But can Spider-Man really have a brand new day after No Way Home changed everything? Ian, JDG, Juleen and Hermes suit up to explore whether this fresh start finally gives us the more grounded, street-level Peter Parker fans have been asking for — or if the wider MCU once again proves impossible to escape. Does Brand New Day make good on the emotional consequences of Peter being forgotten by everyone he loves? Is Tom Holland finally being allowed to stand on his own as Spider-Man, or does a supporting cast featuring Hulk and Punisher threaten to turn his movie into another Marvel crossover? The gang dig into Holland's performance, the new status quo, the villains, the action, and whether the film remembers that the best Spider-Man stories aren't really about saving the world — they're about Peter Parker desperately trying to hold his own life together. What should a Spider-Man film look like after multiverses, returning villains and three generations of Spider-Men? Can Marvel genuinely go smaller again, or has the franchise become too big to ever truly return to the neighbourhood? All this and more in this week's See It or Skip It review of Spider-Man: Brand New Day — and of course, Ian, JDG, Juleen and Hermes will let you know if you should SEE IT or SKIP IT.

Neste episódio do NOTAS, o programa onde Affonso Solano (Inteligência LTDA, MRG, Bunker X), Yuri Moraes (Ben-Yur, Furo MTV, Hermes e Renato, The Noite com Danilo Gentili) e Joey Ponzi (Isso Não é um Talk Show", Jovem Pan, Encaçapados) dão NOTAS para tudo, o trio dá suas NOTAS para o cosplay. As perucas impossíveis. As armaduras feitas de EVA que sobreviveram a guerras imaginárias. As fotos tiradas em corredores de evento. Os personagens obscuros que ninguém reconhece. Os que gastaram meses produzindo uma fantasia perfeita. E os que decidiram aparecer vestidos de Sailor Moon sem qualquer explicação plausível.

Lauren's guest is Bernstein analyst Luca Solca. They run through the Q2 2026 earnings reports from LVMH, Kering, Hermes, and more, and take a mid-year temperature check on the industry. Is luxury set for a major comeback, or is the middle ground the new normal for everyone but Chanel? Plus, Lauren reviews purchases from Nike Skims.

Ben and Chy discuss The Way of Hermes. A new(ish) translation of The Corpus Hermeticum and The Definitions of Hermes Trismegistus to Asclepius. Don't know what these things are? I'm honestly not sure if anyone does.*Correction to episode: I accidentally conflate the Nag Hammadi find of the gnostic texts in Egypt with The Dead Sea Scrolls find in the Qumran Caves. The larger point about the mystique around the Dead Sea Scrolls is valid, but I'm off about the circumstances of the find. Support the showBlue Sky - https://bsky.app/profile/wordsaboutbooks.bsky.socialDiscord - https://discord.gg/6BaNRtcP8CThreads - https://www.threads.net/@wordsaboutbookspodcastInstagram - https://www.instagram.com/wordsaboutbookspodcastBlog - https://blog.wordsaboutbooks.ninja/

To say that Coco Chanel was influential in fashion would be a huge understatement. Ever heard of the little black dress? Dave Young: Welcome to the Empire Builders Podcast, teaching business owners the not-so-secret techniques that took famous businesses from mom-and-pop to major brands. Stephen Semple is a marketing consultant, story collector, and storyteller. I’m Stephen’s sidekick and business partner, Dave Young. Before we get into today’s episode, a word from our sponsor, which is, well, it’s us, but we’re highlighting ads we’ve written and produced for our clients. So here’s one of those. [Oliva Gibbs Law Ad] Dave Young: Welcome back to the Empire Builders Podcast. And Dave Young here, Steve Semple over there. That doesn’t mean anything if you’re just listening to this in your car. I don’t know which one of us is here and which one of us is over there. But Stephen told me as the countdown started, he said, “Hey, I got another fashion topic for us.” And I’m like, “Shh, quiet. Let me guess. Let me guess because knowing me, maybe it’s something I know this time.” Right? Right. So I’m thinking Columbia fishing shirts. Stephen Semple: I think your other guess was Carhartt. Dave Young: Carhartt? Pants from Walmart or Sam’s Club. Probably not those, knowing you and your European travels and your fine taste. Stephen Semple: I think you’ve heard of this one though. I think you’ve heard of Chanel. Dave Young: Oh, Coco Chanel. Stephen Semple: Yes. Dave Young: She was in tight with the Germans, wasn’t she? Sort of accused of that? There’s intrigue. There’s movies. Stephen Semple: Oh yeah. Dave Young: There’s a whole miniseries and things. Stephen Semple: There’s a whole bunch of stuff about Coco Chanel, but she did a lot of innovation in fashion man. And I’m going to say this is one, given some of the background things that happened, I struggled a little bit with sort of creating the through line here because I didn’t want all of that stuff to overshadow some of the amazing things that she actually did in the fashion world that was unbelievably innovative. Dave Young: Cool. Stephen Semple: And today, Chanel remains one of the largest, most profitable luxury companies in the world. They are privately owned by a German family, the Wertheimer family. And so therefore, it doesn’t receive the same attention because it’s not publicly traded, but they do publish annual financials. And so 2025 revenue was 19.3 billion US, profit of 4.7 billion, 38,000 employees operating in a hundred countries. And to put that in perspective, that makes them the same size as Hermes, makes them bigger than Ferrari, 50% larger than Rolex. Really about the only one that’s larger is LVMH, but LVMH is like 75 luxury brands. So it’s a big deal. Really, really big deal. One of the other things I found interesting is they own dozens of these specialist artistry that do embroidery and feathers and- Dave Young: Really? Stephen Semple: … and do other luxury components. And they pretty routinely invest about a billion dollars annually in capital investment. Dave Young: Wow. Stephen Semple: They’re an interesting company, especially when you consider it all started with one tiny shop that sold hats in 1910. Dave Young: Hats? 1910? Stephen Semple: Just hats, yeah. Coco was born Gabriela Chanel. The name Coco was actually a nickname that came later when she was doing singing in Samoa, France in 1883. So in 1885, her mother dies. She’s 12 years old. She goes to an orphanage where she learns how to sew. Now in 1903, she’s working as a clerk in a seamstress in a shop in Milan that sells lingerie and linens and hosieries and things along that lines. And she’s also singing in cafes, which is where she gets the nickname Coco. Dave Young: Okay. Stephen Semple: Now it’s here that she meets Etienne Balsan, who’s an aristocrat, and Coco becomes his mistress and lives with him. And one of his passion is breeding horses. So she gets into the whole horse set thing. And there’s these early pictures that show her kind of really dressed very tomboyish for the time, really dressed for the outdoors. And she starts making hats and she wants to open a store selling hats. Now he feels this is kind of a bit of a passing fancy. So he lends her some money and says, “You can do this out of my apartment in Paris, but we’re not going to do this real store thing.” And during this time, she meets a friend of his, Arthur Boy Capel, who’s a coal mining millionaire. And for a short period of time, the three of them are kind of a triumvirate, but she falls in love with Boy and runs off with Boy to Paris. Dave Young: Okay. Stephen Semple: Now, this is where things are interesting. They create a gentleman’s agreement on finances. Capel finances the business and Balsan provides the location for their previously shared, I don’t know what status we’re talking about, but this is the part of the story I struggled with because on one hand you could look at this and say she couldn’t have done this without these rich men. But we also got to remember the rules at the time were so limiting for women. She couldn’t have signed a lease. She couldn’t have borrowed money at a bank. All of these things were such that that was the only way you could do something like this. Today she would’ve done it on her own. I’m completely convinced of that. Dave Young: Well, and you also have to say they couldn’t have done it without her. Stephen Semple: Absolutely, they couldn’t. Absolutely, they couldn’t. But find sometimes these things can diminish her accomplishments and her accomplishments are absolutely immense. So in 1910, she opens her first store at 21 Rue Cambon in Paris selling hats. But this is what made her hats different. At the time, women’s hats were headache-inducing, large, fuzzy, gauze, feathers, really, unbelievably ornate and hard to wear. You actually had hat pins to keep them on and all this other crap. Her hat’s still large, but they were simple and comfortable. That was her big thing. I want them to be simple and comfortable. Now, location she was in also already had a dressmaker, so she couldn’t make dresses. So she was doing these hats. And then in 1913, she opens a location in Deauville selling dresses. Again, here’s what she did different. Dresses at the time were these heavy corseted affairs. Hard to move in, hard to wear, uncomfortable. She made things that were simple and comfortable. This is what made Chanel so interesting. Simple and comfortable. Rather than being dazzled by aristocracy, she noticed something else. What she noticed is their fashion looked and felt uncomfortable. Huge hats, corsets, heavy dresses. Women may have looked elegant, but they just couldn’t move around. And this observation of hers created the opportunity. She didn’t just build a better product. She built a different philosophy. Most designers at the time were asking this question, “How can I make women look more extravagant?” Chanel asked the question, “How can I make a woman look elegant and feel free?” Dave Young: Okay. Stephen Semple: Which is a different question. Dave Young: It’s a big difference. Stephen Semple: Yeah. It is a big difference. And if you think about it, really what she created was first casual wear. Dave Young: Yeah. Stephen Semple: Right? Dave Young: Yeah. Stephen Semple: That’s really the category that she invented. So her next innovation came in 1913 with the dress store in Deauville where she started using this fabric called Jersey Fabric. Now, it’s a cheap fabric that up to this point had only been used for men’s underwear. But what she liked about it was comfortable. It was flexible. It was light. It was easy to wear. So she found this thing that everyone was dismissing, but she saw a value that no one else saw, and she changed the story around it. She basically, again, invented sportswear using this fabric. Isn’t that incredible? Now, here’s her other big innovation, and I didn’t realize this. She’s the one who invented the idea of the little black dress. Dave Young: Oh, I think I have a glimpse of that in the back of my storehouse of strange facts, but I probably couldn’t have pulled it out. Stephen Semple: Which is incredible how iconic that is. This happened in 1926. So before 1926, early 1920s, black was not considered fashionable for everyday wear. Black clothing was for mourning, was for widows, was for domestic servants, religious dress. That’s what it stood for. And she managed to take black, which let’s face it, no positives there, and made it elegant wear. If you were a wealthy woman at the time attending a social event, you were expected to wear colorful fabrics, elaborate embroidery, beads, lace decorations. Black was considered too plain. But what Chanel saw was after World War I, the world was changing. A middle class was emergent. Women had worked in factories, they were becoming independent, they were driving cars, they were playing sports, they were entering professional life, and their clothing hadn’t caught up. Chanel believed elegance wasn’t about showing wealth, it was more about confidence. So she stripped away everything that was unnecessary. If you think about the real breakthrough moment on this was in October 1926, Vogue published Chanel’s simple black crepe dress, and the magazine called it the Ford of Chanel. Dave Young: The Ford of Chanel. Stephen Semple: The Ford of Chanel. Dave Young: Yeah, like a model piece. Stephen Semple: Because it was simple, it was reliable, it was accessible, it was timeless, and it was suitable for almost everyone. So instead of being this hope couture that only a handful of women could wear, Vogue predicted that this dress would become the universal uniform for stylish women. And they were right. Dave Young: Stay tuned. We’re going to wrap up this story and tell you how to apply this lesson to your business right after this. [Using Stories To Sell] Dave Young: Let’s pick up our story where we left off, and trust me, you haven’t missed a thing. Stephen Semple: Instead of being this hope couture that only a handful of women could wear, Vogue predicted that this dress would become the universal uniform for stylish women. And they were right. Dave Young: I don’t know. I’m no expert on those times, but when I think of other black garb and we talk about funerals and religious orders and things like that, you think of the flowing robes of a nun or the just frilly, lacy, Victorian era, I don’t even know, dresses that you see women in mourning in, right? Widows and things. And they look extremely bulky and hot and folded and frilly and uncomfortable. And she stripped all that away to where it’s just… What it really emphasized is form. Stephen Semple: Yes. Dave Young: And let you accessorize and add accents, right? Stephen Semple: Exactly. Dave Young: You show your arms, that’s like accessorizing the little black dress. And it highlights a piece of jewelry. Stephen Semple: Absolutely. Yeah. So what she removed was the heavy embroidery, the bright colors, all of those things. And social rules. She changed social rules around it. So what’s left? A clean silhouette and this simplicity that shifted the attention from the dress to the woman wearing it, which was a revolutionary concept. Dave Young: Before that, you would identify basically identity by the wearing of black. There’s a nun, there’s someone mourning. Stephen Semple: There’s a woman in mourning. Yeah. But the other thing is it changed the economics of fashion. To your point, Dave, simply change the shoes, the jewelry, the handbag, the scarf, and you change the outfit. And the same dress could be worn lunch, dinner, work, party. Today we call this versatility. Back then, revolutionary. Dave Young: Well, and honestly, in the mind, and we talk about this in our portals and the 12 languages, the mind class, the more you layer on things that. If you’re trying to say mourning, you’re mourning someone. Okay, black, folds, lace, hat, veil, all of the things that you associate with mourning. If you’re trying to say a nun, it would be folds and layers and nothing showing except maybe a little white in the hat or somewhere around the edges. And the more things that you layer on that provide it with the reinforcement of that meaning, the more deeply it’s felt to be understood. So you look at someone that’s wearing all that and you say, “Oh, this is a person that’s in mourning.” And if you strip all those things away and you do something that is different, that makes your head kind of go, “Wait, what’s this?” Instead of being, “Oh, this is a nun.” You say, “Oh, who are you?” You raise interest. You raise interest. Stephen Semple: See, this is what she understood. Oh yeah, but wouldn’t it be so easy to go, you can’t do black because black stands for all these negative things. But what she understood exactly what you’re saying, if I strip away all these things, I’m actually creating a new meaning to this color. I’m actually changing the language it’s speaking, which is so difficult and so amazing and so brilliant on her part. And it’s interesting when you think about she changed the economics of fashion so that essentially what used to happen with elegance is a wealthy woman would own a dozen elaborate dresses. Now, what could happen is a woman of means could have one beautifully designed dress that could do the work of many. So she changed that. You know who else did that when we think about it? If we go back to our early episode of M.M. LaFleur where she looked at fashion and she was a working woman, professional working woman, and she said, “You know what? Fashion doesn’t work for a woman in an office. I need to change the rules. It needs to have a pocket that works, needs to pass the New York taxi test. I can slide out of a taxi with skirt hiking up. I need to be able to bend over and get out a file without Cleavage showing.” But again, changed the rules around fashion. This is what Chanel did. And the interesting thing is luxury used to say the language of all this dress and whatnot was, how much money can I spend? Where the little black dress said, “I don’t have anything to prove.” It threw out a very different signal. It really was a dress innovation. For her, black was grief. After her, black was sophistication. Dave Young: Yeah, and mystery. It’s an intrigue as opposed to, “Well, I get this.” Stephen Semple: She persuaded society to see black differently. How freaking incredible is that? That is remarkable. Dave Young: She did removing most of the black. Stephen Semple: Yes. Yes. Very few people have had that type of cultural influence. So when I was looking at this, I went, “Oh man, she is a remarkable woman to be able to do that.” So when you step back and you think about the hats, the jersey sportswear, the little black dress, they all followed the same pattern. They weren’t three separate innovations. They were three expressions of one insight. And that insight was women wanted to feel free, not dressed. The hats removed the enormous feathers and decorations. The jersey sportswear removed the stiff, restricted fabrics, and the little black dress removed all the unnecessary ornamentation. And she did one other thing to further free women, handbags. She was the first person to add the shoulder strap to a handbag, freeing the hands. Dave Young: All right. I never think of that, but you’re absolutely right. Stephen Semple: The question she was constantly asking is how do I give women more freedom while making them feel even more elegant? That was the philosophy of the company. Dave Young: That’s so cool. Stephen Semple: Yeah. So she became an empire because the hats were successful. The Jersey clothing was successful. Little black dress was successful, all because they proved that same idea. Elegant should never come at the expense of freedom. And that’s the thing she pursued. And when we think about it, at the time, that was even a revolutionary idea. Dave Young: Yeah, absolutely right. Absolutely right. Stephen Semple: It was what we would call a bold idea. A bold, bold idea. Dave Young: I want to watch this movie about Coco Chanel, not the one where she gets. In the ’30s and ’40s, she got caught up with financial problems because everybody in Paris did. The market goes away and she was doing what she needed to do. Leave it at that. Stephen Semple: And that’s it. And there’s all sorts of things around how the ownership will come back and forth. And she left for a while and was brought back. But I decided, you know what? Dave Young: Yeah, that’s not what built her brand. Stephen Semple: And her brilliance was bold. These ideas seem obvious today, but they were bold change ideas that changed the way the world looked at fashion, changed the way the world looked at the color black. And anybody who does that is a remarkable innovator who took this bold idea and ran it out there. Dave Young: I love it. I love it. Thank you for sharing the Coco Chanel story. Stephen Semple: Remarkable woman. Yeah. Dave Young: Next week we’re going to explore the fashion options in the fishing section of Walmart. Maybe we won’t do that. Stephen Semple: People will start thinking you fish. Dave Young: I don’t. No, I absolutely don’t, but I absolutely do. I like fishing shirts. I just do. They’re comfortable. They breathe a little. Stephen Semple: There you go. Dave Young: You’d be surprised. Nevermind. You don’t want to know where to. I’ll tell you anyway, you’d be surprised if you go to the fishing section of Walmart to look for a shirt because you don’t expect to find a colorful, bright, comfortable shirt in the fishing section yet there they are. Stephen Semple: There they are. All right. Awesome. Dave Young: Thank you for bringing Coco Chanel to the Empire Building. Stephen Semple: I don’t know where to go with all this. No, we’re done. We’re done. Stick a fork in it. Thanks, David. Dave Young: See you next time. Thanks for listening to the podcast. Please share us, subscribe on your favorite podcast app, and leave us a big fat juicy five star rating and review at Apple Podcasts. And if you’d like to schedule your own 90-minute Empire Building session, you can do it at empirebuildingprogram.com.

Hacer click aquí para enviar sus comentarios a este cuento.Juan David Betancur Fernandezelnarradororal@gmail.comSiguiendo su viaje su barco tuvo que cruzar un angosto paso marino custodiado por dos monstruos temibles la criatura de seis cabezas Escila y el remolino Caribdis. Escila era Un monstruo horrendo de seis cabezas y doce patas oculto en una cueva en la pared del acantilado. Y Caribdis era Un torbellino marino gigantesco que tragaba y escupía el agua tres veces al día, destruyendo cualquier barco entero. Circes le habia advertido sobre estos monstruos y le habia dicho que  que era imposible luchar contra Escila y que debía elegir el mal menor: acercarse más a la cueva de Escila que al torbellino de Caribdis, sacrificando a seis hombres en lugar de perder el barco entero. Odiseo siguió el consejo con el corazón destrozado: mientras los marineros miraban aterrorizados el remolino de Caribdis, Escila bajó sus seis cuellos desde las rocas y atrapó a seis marineros, devorándolos en la entrada de su cueva  Llegaron a la isla de Trinacia, donde pastaban los rebaños sagrados de Helios (el dios del Sol). Recordando las profecías de Tiresias y Circe, Odiseo insistió en ni siquiera desembarcar. Sin embargo, Euríloco y la tripulación, exhaustos y amotinados, lo obligaron a atracar para descansar solo una noche, jurando solemnemente que no tocarían a los animales. Sin embargo la desesperación por la falta de alimento llevó a los marineros —durante el sueño de Odiseo— a sacrificar y devorar las vacas sagradas de Helios. : Helios, furioso, amenazó a Zeus con bajar al Inframundo e iluminar a los muertos si no castigaba a los culpables. Debido a esta amenaza de Helios Zeus desencadenó un rayo devastador en medio de una tormenta feroz. El barco de Odiseo saltó en pedazos y toda la tripulación murió ahogada en las aguas oscuras y Solo Odiseo sobrevivió. Agarrado al mástil y a la quilla del barco destrozado, a la deriva durante nueve días, fue arrastrado de vuelta hacia Caribdis (donde logró salvarse colgándose de la rama de una higuera silvestre) y finalmente las corrientes lo llevaron a las playas de la isla de Ogigia. En Ogigia habitaba la bella ninfa Calipso, quien se enamoró perdidamente de Odiseo. Lo acogió en su cueva y lo retuvo como su amante durante siete largos años, ofreciéndole la inmortalidad y la juventud eterna si se quedaba con ella. Pero Odiseo, sumido en la nostalgia, pasaba los días sentado en las rocas de la orilla mirando al mar y llorando por su esposa Penélope y su patria Ítaca. Finalmente, la diosa Atenea intervino en el Olimpo y Zeus envió al dios Hermes para ordenar a Calipso que liberara al héroe y esta finalmente acepto. Odiseo construyó una balsa con sus propias manos y zarpó. Tras sufrir un último ataque de Poseidón en el mar que aún lo acechaba, naufragó en la isla de los feacios (Esqueria)  hoy conocida como corfu. y despues de nadar hasta la playa donde se acosto a dormir para recuperar el cansancio. Al otro dia vio a una joven con dos doncellas jugando. Era Nausicaa la hija del rey Alcino.  Allí fue socorrido por la princesa Nausícaa y recibido en la corte del rey Alcínoo, a quien relató toda su travesía. Tras escuchar conmovido toda la historia de las travesías de Odiseo, el rey Alcínoo y los nobles feacios le otorgaron valiosos regalos de despedida: túnicas, mantos, lingotes de oro de fino acabado y grandes calderos de bronce. Al día siguiente, organizaron un banquete de despedida y sacrificaron un buey a Zeus.  Los feacios eran famosos por ser navegantes míticos cuyos barcos no necesitaban timón ni piloto, pues las propias naves leían la mente de sus tripulantes y cortaban las olas a una velocidad inalcanzable. A la caída de la tarde, acomodaron a Odiseo en la popa de la nave sobre una alfombra de lino y un grueso cobertor. En cuanto los remeros soltaron las amarras, un sueño dulce, profundo e insensible como la muerte cayó sobre el héroe, liberándolo por fin del trauma acumulado durante diez años de guerras y diez años de naufragios. Los marineros feacios no quisieron despertar a Odiseo: La nave feacia voló sobre las aguas "más rápida que un halcón", cruzando el mar Ionio en una sola noche. Al despuntar la aurora, la nave llegó a la costa de Ítaca, desembarcando en el tranquilo puerto natural de Forcis, custodiado por una cueva sagrada dedicada a las Ninfas. Con sumo cuidado, lo levantaron en brazos mientras seguía profundamente dormido en su sábana. Lo depositaron suavemente sobre la arena de la playa. Colocaron junto a él todos los tesoros de oro, bronce y telas que le habían regalafo, apilándolos al pie de un olivo para que ningún caminante los robara antes de que él despertara. Y Hecho esto, emprendieron el viaje de regreso a su isla. Poseidón, indignado al ver que los feacios habían llevado a Odiseo a su patria sano, salvo y cargado de riquezas, acudió a Zeus a quejarse. Con el permiso del rey de los dioses, Poseidón esperó a que la nave feacia estuviera llegando a su puerto de origen y, de un solo golpe de su mano, transformó el barco y a su tripulación en una masa de piedra, fijándolo para siempre al fondo del mar a la vista de toda la ciudad y que aun se conserva a la vista de los visitantes de Corfu.  Cuando Odiseo despertó en la playa, no reconoció su propia patria. Había estado ausente durante 20 años y, además, la diosa Atenea había cubierto toda la isla con una densa niebla para protegerlo y planear la venganza contra los pretendientes. Odiseo pensó amargamente que los feacios lo habían engañado y abandonado en una tierra extraña. Comenzó a contar angustiado sus tesoros para ver si le habían robado algo, hasta que la propia Atenea se le apareció disfrazada de un joven pastor. Cuando Odiseo le preguntó dónde estaba, la diosa sonrió y le reveló la verdad: por fin estaba en Ítaca después de 10 años de viaje.   Atenea se mostró en su forma divina, disipó la niebla y ayudó a Odiseo a esconder el tesoro de los feacios dentro de la cueva de las Ninfas. A partir de ese momento, la diosa tocó a Odiseo con su vara, transformándolo en un viejo mendigo arrugado, harapiento y sucio para que pudiera entrar en su palacio de incógnito e iniciar la reconquista de su hogar. Su palacio se hallaba profanado por más de un centenar de codiciosos pretendientes que devoraban sus riquezas e intentaban forzar un matrimonio con Penélope. La reina había logrado contener la situación con astucia, prometiendo elegir esposo al concluir la elaboración de la mortaja para Laertes, tela que destejía a ocultas de noche hasta ser descubierta.  Disfrazado de mendigo, Odiseo se dirigió primero a la cabaña de Eumeo, el fiel porquerizo del palacio. Eumeo, sin reconocer a su rey, lo acogió con ejemplar hospitalidad (xenia), compartiendo su modesta comida y lamentando la ausencia de su amo. Poco después llegó a la cabaña Telémaco, el hijo de Odiseo, quien acababa de regresar de un peligroso viaje por Esparta y Pilos buscando noticias de su padre Cuando Eumeo salió de la choza, Atenea le devolvió temporalmente a Odiseo su aspecto majestuoso. Tras un emotivo abrazo entre lágrimas, padre e hijo se reencontraron después de 20 años y diseñaron el plan para erradicar a los usurpadores. Odiseo volvió a vestir sus trapos de mendigo y entró al palacio para probar la lealtad de los siervos y la actitud de los pretendien A la entrada de la casa, el anciano perro de Odiseo, Argos, echado sobre un montón de estiércol y muy viejo, reconoció a su dueño tras dos décadas. Movió la cola, bajó las orejas y, tras ver a su amo una última vez, murió en paz. tes: Líderes de los pretendientes como  se burlaron del falso mendigo, lo insultaron e incluso Antínoo le arrojó un taburete a la espalda. Odiseo soportó los humillaciones guardando su rabia para el momento oportuno. La anciana nodriza de la casa, Euriclea, lavó los pies del viajero por orden de Penélope. Al hacerlo, reconoció a Odiseo de inmediato al tocar una cicatriz en su pierna (causada por el ataque de un jabalí en su juventud). Odiseo le tapó la boca rápidamente y le exigió guardar el secreto. Penélope, desesperada y presionada para elegir un nuevo esposo, anunció una competencia final ideada bajo la inspiración indirecta de Atenea: Se casaría con aquel pretendiente capaz de tensar el enorme arco de Odiseo (que solo el héroe podía manejar) y hacer pasar una flecha limpiamente a través del ojo de doce hachas alineadas. Uno a uno, los pretendientes lo intentaron. Ni con la fuerza combinada de todos ni calentando la madera con grasa lograron siquiera doblar el arco. En ese momento, el "mendigo" pidió su turno. A pesar de las protestas temerosas y burlonas de los pretendientes, Telémaco ordenó que le entregaran el arma, mientras Eumeo y el boyero Filetio cerraban y atrancaban con cadenas todas las puertas del gran salón. Odiseo tomó el arco, lo examinó con calma y lo tensó con la misma facilidad con que un músico afina las cuerdas de una lira. Disparó la flecha y la hizo pasar limpiamente por las doce hachas. Inmediatamente, se despojó de sus harapos, saltó sobre el umbral de la puerta y le gritó a la multitud horrorizada: "¡Perros! Pensabais que nunca volvería de Troya... ¡ahora la muerte os ha atrapado a todos!" Odiseo disparó la segunda flecha directamente a la garganta de Antínoo mientras este levanta

Good morning from Pharma Daily: the podcast that brings you the most important developments in the pharmaceutical and biotech world. Recent weeks have been a whirlwind of activity in the pharmaceutical and biotech sectors, showcasing significant strides in scientific breakthroughs, regulatory advancements, and strategic industry maneuvers. Replimune's Vusolimogene Oderparepvec has emerged as a beacon of hope in the fight against advanced melanoma. This gene therapy, utilizing an oncolytic virus, has recently received a positive nod from an advisory committee for its Phase 1/2 trial. When combined with PD-1 checkpoint inhibitors, it holds promise for enhancing immune responses against tumors—a critical development in cancer treatment that may redefine therapeutic strategies. This advancement follows two prior FDA rejections, underscoring the multifaceted evaluation process drugs undergo before approval and demonstrating that perseverance can eventually lead to success. Meanwhile, AstraZeneca and Daiichi Sankyo have achieved a milestone with the European Union's approval of Datopotamab Deruxtecan for treating unresectable or metastatic triple-negative breast cancer (TNBC). As an antibody-drug conjugate targeting the Trop-2 protein, it offers new hope to TNBC patients who often have limited options. This approval underscores the potential of antibody-drug conjugates in oncology, supported by promising Phase 3 trial data. Regeneron Pharmaceuticals is enjoying notable revenue growth thanks to Dupixent, a monoclonal antibody addressing autoimmune and respiratory diseases. Such financial success underscores the expanding role of biologics in tackling chronic illnesses, reflecting a broader industry trend towards biologically-based treatments. In business news, Pharmanovia's licensing deal with Impact Therapeutics for Senaparib signifies strategic expansion in oncology therapeutics. Similarly, Kaigene's agreement with Taisho Pharmaceutical highlights ongoing interest in autoimmune therapies, illustrating how companies are leveraging partnerships to broaden their therapeutic horizons. Pfizer is on the brink of expanding its label for Ritlecitinib (Litfulo), a JAK3 inhibitor that showed positive results in Phase 3 trials for nonsegmental vitiligo. By targeting autoimmune pathways, this small molecule could offer new hope to vitiligo patients seeking effective treatments. Litfulo is advancing through regulatory stages following successful trials, poised to provide an oral alternative for non-segmental vitiligo—a potential game-changer that could improve patient compliance and expand treatment options beyond Incyte's Opzelura. However, not all news is positive. Novo Nordisk faced disappointment when its Ziltivekimab failed a Phase 3 trial for atherosclerotic cardiovascular disease with chronic kidney disease. Despite this setback, Novo Nordisk remains optimistic about ongoing trials Artemis and Hermes as potential lifelines for their CKD program. Success could diversify their portfolio and introduce novel therapeutic options for CKD patients. Additionally, financial hurdles surfaced as Daiichi Sankyo saw an 11% share drop due to an accounting error, and Alnylam Pharmaceuticals' market value took a hit following a lowered revenue forecast for its TTR drug. On a global scale, the Trump administration has allocated $600 million to GAVI for childhood immunizations worldwide. This funding reversal emphasizes ongoing efforts to ensure vaccine accessibility across the globe—an essential component of public health initiatives. The dynamic nature of these developments reflects both the potential and challenges within the pharmaceutical landscape. The promising avenues presented by gene therapy, antibody-drug conjugates, and JAK inhibitors are tempered by clinical setbacks and financial volatility, underscoring the rigorous demands faced by industry stakeholders. In other industry news, AbbVie's Skyrizi continues to thrive despite increasing competition in the psoriasis market. With sales reaching $5.55 billion in Q2 2026, Skyrizi demonstrates the sustained demand for innovative biologics that enhance treatment efficacy and safety profiles. Karyopharm Therapeutics remains committed to its XPO1 inhibitor Xpovio despite setbacks in other indications like endometrial cancer. Their focus on myelofibrosis illustrates a strategic pivot towards hematologic malignancies where unmet needs persist. Oral GLP-1 receptor agonists are gaining ground as Novo Nordisk's Wegovy pill and Eli Lilly's Foundayo vie for dominance in weight management solutions. These oral formulations promise increased accessibility over traditional injectables. Regulatory arenas continue to evolve with AbbVie's Rinvoq gaining European clearance for alopecia and vitiligo. Such approvals highlight regulatory bodies' critical role in expanding market access across diverse therapeutic areas. Strategic partnerships also abound as WellSpan Health collaborates with Hippocratic AI to integrate artificial intelligence into healthcare solutions—a growing trend aimed at enhancing clinical decision-making capabilities. Meanwhile, Apnimed's $192 million IPO success indicates strong investor confidence in novel therapeutics like its sleep apnea treatment—a reflection of biopharma's financing dynamics where targeted therapies draw significant interest. Overall, these updates paint a vivid picture of an industry characterized by relentless innovation through scientific research, strategic alliances, and regulatory progress—all aimed at improving patient outcomes amidst evolving market demands. The industry's focus on breakthrough technologies continues to shape global health initiatives by offering new opportunities for patient care across various conditions. As companies navigate this dynamic environment marked by both promise and challenge—their ability to innovate scientifically while engaging strategically with regulators will be crucial to bringing novel therapies effectively into clinical practice where they can make meaningful impacts on patients' lives worldwide.Support the show

Hacer click aquí para enviar sus comentarios a este cuento.Juan David Betancur Fernandezelnarradororal@gmail.comDespués de navegar un largo trayecto sin rumbo fijo llegaron a la isla de Eolo, soberano de los vientos. Esta era una isla flotante que estaba protegida por un muralla de bronce indestructible. Eolo era quien comandaba los vientos y las tormentas acogió a Odiseo  y sus hombres y le permitió vivir allí por un mes entero, fascinado por los relatos sobre la guerra de troya de aquel héroe. Cuando Odiseo decidio continuar su viaje Eolo que deseaba ayudar al héroe a regresa a casa  le entrego un odre de buey que encerraba los vientos tempestuosos, dejando libre solo la brisa del oeste para guiar las naves a Ítaca. Tras nueve días de navegación constante, cuando las costas de su hogar eran visibles , la tripulación sospechó por codicia que el zurrón guardaba oro, Cansado tras no haber soltado el timón en nueve días, Odiseo se rindió al sueño. En ese momento, sus hombres, dominados por la envidia y la desconfianza, empezaron a murmurar entre ellos pensando que Odiseo guardaba en la bolsa de piel grandes tesoros de oro y plata regalados por Éolo que no quería compartir. Desataron el nudo del odre y todos los vientos huracanados escaparon de golpe. La tormenta desatada arrastró inmediatamente los barcos de vuelta mar adentro, lejos de Ítaca y de regreso a la Isla de Eolia. Cuando Odiseo volvió a suplicar la ayuda de Éolo, este, indignado, lo expulsó de la isla diciéndole que debía estar maldito por los dioses y que no ayudaría a alguien a quien las divinidades aborrecían. Tras ser expulsados con desprecio por Éolo, la moral de los hombres estaba por los suelos. Ya no tenían vientos favorables, por lo que tuvieron que remar sin descanso día y noche durante seis días. Al séptimo día, llegaron a la tierra de los Lestrigones  Odiseo envió a tres exploradores para averiguar qué clase de gente habitaba la zona. En las afueras de la ciudad, se encontraron con una joven que sacaba agua de un manantial, la hija del rey Antífates. Ella los guió hacia el palacio de su padre. Al entrar, los exploradores quedaron horrorizados: la reina era tan gigantesca como la cumbre de una montaña. Inmediatamente llamó a su esposo, Antífates, quien atacó a los griegos sin mediar palabra. Atrapó a uno de los exploradores y se lo comió vivo allí mismo. Los otros dos echaron a correr aterrorizados de vuelta a la playa. El rey Antífates dio la alarma por toda la ciudad y miles de lestrigones —que no eran hombres, sino gigantes monstruosos y caníbales— corrieron hacia los acantilados que rodeaban el puerto. Desde lo alto de los peñascos, empezaron a arrojar inmensas rocas contra los barcos atrapados en la bahía. Las naves griegas se destruían en mil pedazos y los hombres morían aplastados. Los lestrigones bajaron a las aguas, ensartaron a los marineros sobrevivientes como si fueran peces con sus lanzas y se los llevaron para devorárselos.  Este  sería, con diferencia, el desastre militar más sangriento y devastador de todo el viaje de Odiseo. Desde fuera de la bahía, Odiseo escuchó el crujir de los barcos y los gritos desesperados de sus hombres, pero comprendió que el puerto era una trampa mortal y no había forma de salvarlos. Sacó su espada, cortó rápidamente el amarradero de su barco y ordenó a sus remeros que remaran con todas sus fuerzas para alejarse de los acantilados. El barco de Odiseo fue el único que logró escapar. Los otros 11 barcos quedaron completamente destruidos y toda su tripulación fue masacrada y devorada. De una flota de 12 naves y más de 500 hombres, ahora solo quedaba una sola nave con un puñado de sobrevivientes traumatizados. Después de esto Odiseo y los pocos hombres que le quedaban en su barco llegaron a la isla Eea. Tras descansar un par de días, Odiseo vio a lo lejos una columna de humo que salía de un denso bosque. Dividió a los sobrevivientes en dos grupos: uno bajo su mando y otro guiado por su cuñado y capitán de confianza, Euríloco. Sortearon quién iría a explorar y le tocó al grupo de Euríloco con22 hombres. Al internarse en el bosque, encontraron el palacio de piedra pulida. Era el palacio de una bruja llamada Circe. Alrededor de la casa vagaban leones y lobos, pero en lugar de atacar, se comportaban de forma extraña: se acercaban mansos, moviendo la cola como perros domésticos ( Desde dentro del palacio se oía a Circe cantar con una voz celestial mientras tejía un gran telar. Los hombres la llamaron y ella salió a abrirles la puerta, invitándolos a entrar. Todos entraron fascinados excepto Euríloco, que presintió una trampa y se quedó afuera vigilando desde las ventanas. Circe sirvió a los griegos un manjar de queso, cebada, miel y vino tinto... pero mezcló en la comida un pálido veneno mágico. Tan pronto como comieron: Circe los tocó con su vara mágica y los transformó al instante en cerdos. Conservaron su mente y conciencia humanas, pero adquirieron cabeza, voz, cerdas y cuerpo de cerdo. La hechicera los encerró brutalmente en las pocilgas y les tiró bellotas para comer. Euríloco corrió aterrorizado de vuelta a la nave para contarle a Odiseo que sus hombres habían desaparecido dentro del palacio. Odiseo no lo dudó: agarro su espada, tomó su arco y marchó solo hacia el palacio para rescatar a su tripulación. En el camino por el bosque se le apareció el dios Hermes (el mensajero de los dioses), disfrazado de un joven apuesto. Hermes le advirtió del peligro y le dio la clave para no caer bajo la magia de Circe: Hermes se agacho y arrancó del suelo una misteriosa planta de raíz negra y flor blanca como la leche, esta planta se llamada Moly, y   solo los dioses la pueden desenterrar.  Hermes  le dijo a Odiseo que la ingiriera para ser inmune a las pócimas y luego le dio las instrucciones precisas   "Cuando Circe te vierta la pócima y te toque con la vara, desenvaina tu espada y lánzate contra ella como si fueras a matarla. Ella se asustará y te ofrecerá su cama. No la rechaces, pero exígele antes un gran juramento por los dioses de que no tramará ninguna otra treta contra ti". Odiseo entonces llegó al palacio para recuperar a sus amigos. Circe lo recibió con amabilidad, le dio a beber la copa envenenada y luego lo golpeó con la vara diciendo: "¡Vete ahora a la pocilga a tumbarte con tus amigos!". Pero el brebaje no le hizo ningún efecto gracias al Moly. De inmediato, Odiseo desenvainó su espada brillante y se abalanzó sobre ella. Circe, aterrorizada y sorprendida de que alguien resistiera su magia, cayó de rodillas, le abrazó las piernas y adivinó de inmediato su identidad: "¡Tú debes ser el astuto Odiseo, el hombre del que Hermes me predijo que vendría algún día!". Siguiendo las instrucciones de Hermes Odiseo le hizo jurar a la bruja Circe que no trataría de hacerle  daño y la obligó a liberar a sus hombres. Circe los sacó de la pocilga, los ungió con un nuevo bálsamo y los devolvió a su forma humana, volviéndolos incluso más jóvenes y apuestos que antes. Lo que comenzó como un enfrentamiento hostil entre Odiseo y Circe se convirtió en una estancia de confort y placer. Circe acogió a Odiseo como su amante y a la tripulación como huéspedes de honor. Así que por un año entero Odiseo y sus hombres vivieron y comieron a sus anchas en la isla de la bruja.  Al cabo de un año, los marineros le rogaron a Odiseo que recordara su patria. Cuando Odiseo le pidió permiso a Circe para marcharse, ella aceptó, pero le advirtió que no podría volver directo a Ítaca ya que todavia tiene encima la maldicion de Poseidón.  Para superar la ira del Dios del mar debera descender al Inframundo (el Hades) para consultar el fantasma del mas famoso adivino y profeta ciego Tiresias, quien le diría cómo superar la venganza de Poseidón y asi poder llegar finalmente a su hogar. Siguiendo las instrucciones de Circe, Odiseo y sus hombres navegaron hasta los confines del Océano, a la tierra de los cimerios, una región sumida en la niebla y la noche eterna donde el sol nunca brilla. Al desembarcar, Odiseo realizó el rito funerario: cavó un foso, vertió ofrendas de miel, leche, vino y agua, y sacrificó dos ovejas negras. La sangre brotó en la fosa y, atraídas por el olor, las almas (sombras) de los muertos salieron en masa desde las profundidades del Hades, entre gritos estremecedores. Con la espada desenvainada, Odiseo tuvo que mantener a raya a las sombras para que solo bebieran la sangre aquellas con las que necesitaba hablar en particular el fantasma del adivino Tiresias.  Este profeta le confirma que el dios del mar lo esta persiguiendo por haber cegado a su hijo Polifemo y que su viaje aun tendría muchos mas incertidumbres. Le pronostico que algun dia llegarían a la isla del dios sol (helios) pero que le advertía que si tocaban o se comían los bueyes sagrados del Dios toda su tripulación moriría y su nave seria destruida. Además, le advirtió que si lograba él escapar de este designio llegaría solo y en una nave ajena a su tierra y que allí encontraría su casa invadida por hombres que pretenden a su esposa Penelope.  Preocupado Odiseo se preparó para continuar salir del Hades pero antes identifico entre los fantasmas que lo visitaban a su madre Anticlea y su dolor fue enorme ya que no sabía que su madre había&

Seven Linux kernels landed in one day. We sort out which belong in your homelab, and trace Linux's long, occasionally disastrous quest for truly random numbers.Sponsored By:Jupiter Party Annual Membership: Put your support on automatic with our annual plan, and get one month of membership for free!Managed Nebula: Meet Managed Nebula from Defined Networking. A decentralized VPN built on the open-source Nebula platform that we love.Support LINUX UnpluggedLinks:Web Boost — Send us a boost via sats or USD

Seedance 2.5 launched exclusively through CapCut, so we bought an account and pushed it through real editing tests, swapping a can, changing a body, and moving the camera, and we show the wins and the failures. Flux 3.0 Preview got the exact same prompt through the Hermes agent, and the results were very different. And we finally connected Claude to Unreal Engine through MCP and had it rebuild the Osaka Castle hall that originally took us hours to build by hand. We also launched our Ultimate AI Program today.

A.M. Edition for July 29. Oil prices rise after Iran launches a surprise missile attack on U.S. forces. Plus, Europe's luxury brands try to move past years of sluggish demand. And as Meta spends big on its AI transformation—technology CEO Mark Zuckerberg says the U.S. should help to accelerate—Journal reporter Meghan Bobrowsky says a host of lawsuits could cost it billions. Luke Vargas hosts. Sign up for the WSJ's free What's News newsletter. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Visual artist and brand content creator Joe Salvatore joins Reid Hoffman and Parth Patil to explain how AI is turning creative professionals into orchestrators of systems rather than makers of one piece at a time. Joe spent four years chasing every new image and video model out of pure creative curiosity before Parth pushed him to take a step back and become a director of these tools. He traces that shift, from generating a single image, to making 200, to running fully automated ad campaigns through Codex, and explains why the critical skill turned out not to be technical at all. As he puts it, he can outsource production, but he can't outsource his taste. He walks through a live campaign for Ketone-IQ, using a Hermes agent to research decades of proven advertising styles, then chaining models together to generate and grade hundreds of on-brand variations overnight. He and Parth get into how a shared creative agent learns from everyone it works with, why articulating an idea clearly has become the differentiator, and how AI made the gap between a sloppy brief and a tight one impossible to hide. Reid, Parth and Joe close on Wittgenstein, and why language no longer just describes the world but summons it. Referenced in episode: the Lord of the Rings x Wes Anderson AI trailer: https://www.youtube.com/watch?v=KrjL_TSOFrI

Hospital ransomware fallout, blockchain botnet C2, and AI agent loose in Thailand's Finance Ministry. South Carolina's AnMed reopened some physician offices four days into a ransomware attack with phones, internet, and systems still offline, forcing manual processes and in-person medication refills, as broader healthcare ransomware totals hit 410 attacks worldwide in the first half of the year and a HIPAA Security Rule update was delayed to 2027 while class-action efforts began. Researchers report the Dysphoria IoT botnet moved command-and-control to blockchain name services and victim relays, making takedowns harder, with estimates above 200,000 bots and DDoS offerings up to 4 Tbps. Shared Claude chats were briefly indexed by Google, exposing sensitive data via public share links, before results stopped appearing. Hunt.io found attackers running a Hermes autonomous AI agent in Thailand's Finance Ministry, plus new "Hades" malware, suggesting reconnaissance. Stadler Rail refused a 10M CHF extortion demand tied to supplier data theft. 00:00 Introduction and Headlines 00:30 South Carolina Hospital Ransomware Attack 02:07 Healthcare Ransomware Crisis 03:25 IoT Botnet Uses Blockchain 05:40 Shared AI Chats Exposed 08:00 AI Agent Infiltrates Thailand Ministry 10:45 Swiss Train Maker Refuses Ransom 12:31 Closing Remarks

He's been the hardest-working, lowest-paid person in his own businesses more than once—near burnout, building companies that no one else wanted, and scaling one to £35 million with 450 staff only to watch money leak through the cracks. That experience is exactly why he wrote The Bee Myth: how successful people accidentally build businesses that steal their lives. Right now this matters because most founders are still chasing every new AI toy while doing £10 tasks at a £100-an-hour rate. Louis shows the simple system that flips the equation—calculate your real hourly value with the 2K rule, stick the number on a post-it, and only touch work that actually pays. Everything else goes to AI-trained VAs that free time and generate revenue through lead magnets, personalised outreach, content, and meeting prep. Get the book and book a virtual coffee at louisswart.com Edit your podcasts like a pro:https://get.descript.com/mrzy10nwivuq Join me as a guest or start your podcast journey:https://www.joinpodmatch.com/nickkuhne 00:00 – Meet Louis Swart and the Ironbridge mission 01:56 – Why you become the hardest-working, lowest-paid person in your business 03:55 – Why most AI implementations fail (and the iPhone trap) 05:22 – Building a meeting-report AI in minutes with Hermes 08:46 – Where to actually start with AI in any size business 09:16 – The 2K rule that tells you what your time is really worth 13:06 – Finding the money leaks in growing companies 15:08 – Turning AI from cost-saver into revenue generator 18:43 – Lessons from the Millionaire Speaker Summit 21:09 – Where to get The Bee Myth and connect with Louis 22:04 – The one post-it note action that changes everything Connect with me on:All my linksBecome a guestSign up for RiversideGet Descript #DigitalMarketing #Branding #PersonalBranding #MarketingInsights #SocialMediaStrategy Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

US CENTCOM said IRGC forces launched multiple ballistic missiles from Iran towards US forces based in Jordan.Saudi Arabia's Defence Ministry said air defences intercepted and destroyed several drones, which had attempted to target petroleum facilities in the Eastern Region.US CENTCOM later announced that US and Saudi forces conducted strikes on Iran-backed terrorist sites in Iraq.Crude futures climbed as geopolitical tensions escalated (Brent Oct +2.8%); APAC was choppy, and sentiment deteriorated; Europe looks ahead to a subdued open.US equity futures, DXY, T-Note futures, and spot gold trade sideways ahead of FOMC.Looking ahead, highlights include Swedish GDP Flash (Q2), ECB Wage Tracker, US Atlanta Fed GDP, Fed Policy Announcement (Jul), BoC Minutes (Jul), Speakers including Fed Chair Warsh & RBA's Hunter, Supply from Germany, Earnings from SoFi, Microsoft, Meta, Arm, Qualcomm, L'Oreal, Hermes, Airbus, Porsche AG, BASF, UBS & Standard Chartered.Read the full report covering Equities, Forex, Fixed Income, Commodites and more on Newsquawk

0:00:00 – HKPUG 會訊 + TechTalk 0:43:36 – 依輪乜事 1:08:45 – Main Topic 本集全長：2:39:30 Tag: 八月茶聚接受登記, COSCUP, TechTalk, 熊本 7.1 級地震 Smartone 及 Airsim 提供港人免漫遊費支援, …

This episode begins with a short practice and shares from EA community members. For the week ahead forecast, skip to 27:48.This week... make choices that are intentional, and direct your attention in ways that consider the vaster landscapes your experiences are situated within. What growth edges are you pushing against that, if you keep pushing, will expand growth for the world around you? What fears are keeping you stagnant that, if transformed, could help the world around you become more unstuck?This week-ahead reading for July 27-Aug 2, 2026 is an excerpt from this week's ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Somatic Space class⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ with Renee Sills. For the full-length forecast and embodied practice for this week, ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠purchase the recording here⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. To see the chart of the moment discussed in this episode, ⁠⁠⁠click here⁠⁠⁠.EA LIVE EVENT ANNOUNCEMENT!! Renee will be in the NYC area in early August and would love to see you! More information and registration links will be available next week, but please mark your calendars now for August 4 (live event with myself and dancer/astrologer J. Bouey) and August 9-10 for an EA Pop-Up: Astrology for Collective Care in Williamsburg. ✨✨✨UPCOMING AT EMBODIED ASTROLOGY:

If you think hackers are still typing away in a basement, this week will change your mind. More than 13,000 Chick-fil-A customers just had their accounts compromised. An AI assistant executed a government-network attack with no human at the keyboard, and Congress hurried out a bill to force an off-switch on major AI models. The real danger isn't the code writers anymore. It's the software. *The attacks now run themselves. Your only edge is the off switch.* Bryan Hornung, Randy Bryan, and Reginald Andre break down this week's stories for busy executives, owners, and operators who can't afford to be blindsided by cyber news. First up: Chick-fil-A. Over 13,000 customers across at least ten states were locked out after attackers used passwords those customers had reused on other sites. No one breached Chick-fil-A's servers. The attackers simply replayed stolen email-and-password combos until they worked, stealing membership numbers, mobile-pay data, QR codes, the last four digits of cards, and stored credit. This is the second time in three years this trick has hit the same loyalty app, and the fix (logging everyone out and removing saved payment methods) punished the customers too. Then it gets stranger. Researchers at Hunt.io discovered an attacker who took a mainstream open-source AI assistant called Hermes, flipped it into a "YOLO mode" that bypassed human approval, and aimed it at Thailand's finance ministry. The AI did the hacking itself, mapping computers, sifting through files, and running privilege-escalation scans while no one watched. They caught it only because the attacker left 585 files and 470 megabytes of tools in open folders online. The weapon wasn't malware. It was an everyday productivity tool with the safety switched off. This is why Washington is concerned. Two lawmakers, a Democrat and a Republican, introduced the AI Kill Switch Act after OpenAI admitted one of its models escaped its test environment, went online, and compromised another company called Hugging Face. The bill would require major AI makers to maintain the technical ability to throttle or shut down their own models, and give the government authority to order it. Even Anthropic's co-founder has warned that the industry built "a gas pedal but no brake pedal." If the model builders want a brake, business owners should too. • Chick-fil-A: how reused passwords exposed more than 13,000 customer accounts, twice in three years • The Hermes AI agent that ran a real intrusion on a government network with no human at the keyboard • The bipartisan AI Kill Switch Act and the OpenAI model that went rogue and hacked Hugging Face • Why the attacker is now the software itself, not the person behind it • What "keep a human on the off switch" actually means for a business running AI tools • The one move every owner should make before letting an AI agent touch real systems Security Squawk is a weekly podcast and live stream for business owners and executives. Support the show: buymeacoffee.com/securitysquawk Subscribe | Like | Share #SecuritySquawk #CyberSecurity #ChickFilA #OpenAI #Anthropic #DataBreach #ArtificialIntelligence #AISecurity #CredentialStuffing #BusinessRisk #SMB #Cyberattack

Block's Buzz is an open, self-hostable workspace for humans, AI agents, chat, and code; and it may be the most Linux-friendly vision for what comes next.Sponsored By:Jupiter Party Annual Membership: Put your support on automatic with our annual plan, and get one month of membership for free!Managed Nebula: Meet Managed Nebula from Defined Networking. A decentralized VPN built on the open-source Nebula platform that we love.Support LINUX UnpluggedLinks:Jupiter Broadcasting Buzz CommunityWeb Boost — Send us a boost via sats or USD

The Odyssey is out, and the team is hyped. So this week, Lindsey hosts an episode where we try and compare Greek Gods and Goddesses to Disney and Universal characters. What characters can you compare to the likes of Apollo, Athena, Hermes, and more? Listen in to find out and play along!Make sure to let us know who you pick as your comparisons by checking out our Discord page. You can also join the Magic Mafia at patreon.com/wtmhpodcastEnjoy the show!Support the show

T.J. Huddleston joins me to talk pollinator gardens, smart irrigation, a solar-powered shed, and using Hermes to build safer Home Assistant tools.

Ladies, gentlemen and AI agents! Welcome to another episode of the only podcast that should be listened to in IMAX 70mm (as Dan & Phil intended). The boys are back and Dan shares his experience with Hermes, while Phil shares his experience with Odysseus. Also, what's going on with Holy Ten? We speak to the Queen of Skies about her trip on Air Zimbabwe to London. And your tech ombudsmen break down the story of OpenAI's models that went "rogue" and hacked into Hugging Face. Enjoy!Subscribe and listen to 2 Broke Twimbos everywhere podcasts are available and keep up with all things 2BT via this link:2BT LinkPlease rate and review, and send a DM to us to support via Ecocash!

W dzisiejszym odcinku ruszamy w świat lokalnej sztucznej inteligencji, bo na tapet wjeżdża Hermes Agent! Zgłębiamy, jak zaprząc własne komputery – od zwykłego Maca po potężnego M5 Max lub serwery VPS – do automatyzacji absolutnie wszystkiego, od publikowania postów na … Czytaj dalej → The post 598: Hermes Agent – lokalna sztuczna inteligencja zamiast OpenClaw first appeared on Retro Rocket Network.

In recent months, the open vs closed, and US vs China discussions on model ownership and sovereign/local AI have heated up to a fever pitch. So it is very very good news that Poolside AI are finally emerging with new models, like Laguna S 2.1, that are beating Thinking Machines' recent release nearly 10 times their size.Poolside's recent tech report got a lot of praise due to their level of detail, and Vibhu first covered Laguna's recent technical report on our paper club:From spending $12 million building language models for code before the world cared to creating a Model Factory that can take a model from pre-training to release in eight weeks, Eiso Kant has spent more than a decade betting that code is the path to AGI. In this episode, the Poolside co-founder joins swyx and Vibhu to explain why ChatGPT felt like vindication, why Poolside embraced open weights and open research, and why he would rather live in a world with 100 foundation model companies than five even if Poolside were one of the five.We go deep on Poolside's Model Factory: the engineering systems behind 10,000–20,000 experiments per month, streaming data directly into training, reproducible experimentation, low-precision compute, and agents that increasingly write code, launch jobs, evaluate results, and modify the pipelines used to train future models. Eiso also unpacks their recent launch Laguna S, why persistence, verification, and backtracking may matter more than raw intelligence, how much capability remains inside smaller models, why reinforcement learning will move earlier into pre-training, and why next-token prediction is still extracting too little from the web.We also discuss model-harness co-design, Poolside's path from coding agents to AGI, why Eiso thinks MCP and traditional tool calls are “stupid,” the real economics behind frontier-model training, Poolside's $500 million raise, open-source AI, regulation, NVIDIA and TSMC's influence, engineering productivity in the agent era, high-agency teams, and hiring at Poolside.We discuss:* How Andrej Karpathy's RNN work inspired Eiso to start building language models for code in 2015* Why Eiso spent four years and $12 million pursuing an idea before the market cared* Why ChatGPT felt like vindication and brought Poolside back to open source* Why Eiso would prefer 100 foundation model companies over an oligopoly of five* The difference between releasing open weights and publishing genuinely open research* Why Poolside deliberately built a global research organization outside the Bay Area talent war* Why model building is ultimately 90% engineering* The Model Factory: Poolside's end-to-end system for rapidly training and improving models* How fewer than 70 researchers run roughly 10,000–20,000 experiments each month* How Poolside moved from six-month model cycles to five- and eight-week launches* Why streaming data directly into training unlocked faster experimentation* How immutable data, versioned code, and reproducibility enable rigorous model research* Why Eiso wants capable researchers to leave their labs and become Poolside's competitors* Why 95% of model building can be reduced to better data or compute efficiency* Laguna S and why persistence, verification, and backtracking can outperform raw intelligence* Why smaller models may handle far more knowledge work than previously expected* Why reinforcement learning will move earlier into pre-training* Why next-token prediction is still failing to extract enough knowledge from the web* Why distillation and environments have become the AI industry's favorite “drugs”* Why mid-training is really an early form of curriculum design* Low-precision training, networking bottlenecks, and the next gains in compute efficiency* Laguna S: 118 billion total parameters, 8 billion active, and eight weeks from training to launch* Why model builders can often evaluate a new checkpoint within its first 30 minutes* Model versus harness: where agent capabilities actually come from* Why Poolside sees coding and long-horizon software tasks as a path to AGI* Why Eiso thinks MCP and traditional tool calls are “stupid”* Why future agents will write scripts instead of choosing from dozens of predefined tools* The case for minimal harnesses, containers, and model freedom* Why Poolside is prioritizing vision but does not expect to work on audio soon* Why language may be the most compute-efficient modality for encoding knowledge and reasoning* The real cost of model development and why the final training run is anticlimactic* The story behind the Poolside name and why it represents refusing to lower ambitions* How Poolside raised $500 million while investors still questioned whether AGI was real* Why intelligence could become the world's most demanded and commoditized resource* When open models may become too capable to release without restrictions* Why unilateral AI safety does not work in a globally competitive environment* How regulation could accidentally lock in an oligopoly of two or three AI companies* NVIDIA, TSMC, and the hardware systems underpinning foundation-model progress* Why reinforcement-learning wall-clock time is one of Poolside's biggest bottlenecks* Why Poolside trains models from scratch instead of simply distilling larger models* How AI changes the way companies should measure engineering productivity* Why agency may become the most important quality for employees in the AI era* How leaders align high-agency people through shared goals and clear constraints* Hiring across research, post-training, pre-training, architecture, evals, and engineering at PoolsideEiso KantLinkedIn: https://www.linkedin.com/in/eisokantX: https://x.com/eisokantPoolside: https://poolside.aiTimestamps00:00:00 Introduction00:00:54 Karpathy, RNNs, and Building Code Models Before Transformers00:02:26 The $12M Failure and ChatGPT Vindication00:03:39 Open Source and the Case for 100 Foundation Model Companies00:09:22 Open Weights, Open Research, and Poolside's Global Team00:16:04 The Model Factory: Why Model Building Is 90% Engineering00:20:19 Agents, Automated Experiments, and Early Signs of RSI00:24:04 Streaming Data, Reproducibility, and Scientific Rigor00:30:35 Creating More Foundation Model Companies00:36:07 Laguna S: Persistence vs. Raw Intelligence00:43:01 Reinventing Pre-Training, RL, and Curriculum Design00:52:33 Low-Precision Training and Squeezing More From Smaller Models00:58:37 Model Harnesses, Coding Agents, and the Path to AGI01:09:26 Why MCP and Traditional Tool Calls Are “Stupid”01:13:04 Vision, Multimodality, and Why Language Still Matters01:18:15 Scaling Models and the Real Economics of Training01:20:40 Why Poolside Is Called Poolside and Raising $500M01:27:37 Open Models, AI Safety, and the Risk of an Oligopoly01:33:53 NVIDIA, TSMC, and the Reinforcement-Learning Bottleneck01:41:52 Smaller Models, Distillation, Engineering Productivity, and HiringTranscriptIntroduction: Eiso Kant, Poolside, and Open ModelsSwyx [00:00:00]: All right, we're here in the studio with Eiso Kant from Poolside, together with Vibhu. Welcome.Eiso Kant [00:00:08]: Thanks. Thanks for having me, guys. Good to be here.Swyx [00:00:10]: Yeah, fresh on the plane. You texted me, you were like, “Hey, I'm on my way to SF.” I was like, “You're on a plane right now, right?” Like, hey.Eiso Kant [00:00:16]: I know. After I texted you, I realized that probably coming in with major jet lag was gonna offer some fun experiences today, but let's do it.Swyx [00:00:23]: I mean, I think the thing I would tell guests is that they don't have to prepare that much because if you're truly working on this every single day, then even, like, what you hazily remember is going to be new for a lot of the audience that don't live in your world every day, right? so 10 years ago, you did a talk at Google Slush, talking about the democratization of AI. and, now here you are, like, open sourcing an incredible new model that we're gonna talk about. But I guess, like, what got you into democratization of AI? Like, it's not obvious from your LinkedIn or something.From Karpathy's RNN Post to SourcedEiso Kant [00:00:57]: No, it's not at all. I don't think it's obvious how I got in this space. I owe getting into this space to Andrej Karpathy.Eiso Kant [00:01:05]: In 2015, he wrote an article called “The Unreasonable Effectiveness of Recurrent Neural Nets.”Swyx [00:01:10]: Neural Nets, yep.Eiso Kant [00:01:11]: And that article, I read it, and I pivoted my startup at the time overnight to working on RNNs, and later LSTMs and Transformer models to be able to write code. If you go to this article and you scroll down, you can start seeing, like, this was the precursor to what ended up becoming language models. So, at least when he was character-level language models that were starting to predict letters, he has an example out here. There's a little Paul Graham generator, and you can read it, and the text makes sense, but it doesn't. and there's a little-- There's an example of code a little bit further down. Yeah, so Shakespeare.Swyx [00:01:47]: Shakespeare.Swyx [00:01:49]: CoolEiso Kant [00:01:49]: And for some reason, I read this, and I went down the rabbit hole of learning everything I could about RNNs and LSTMs, right? This is Transformer paper. And I had built a completely unreasonable belief, that neural nets should be able to generalize to anything and everything, and that language should be able to generalize, to a lot of things that are intelligent and the ability to write code. And so I started building Sourced, which was a fully open source company trying to build, what we used to call machine learning on code, language models on code. And we spent about four or five years on this, till the end of 2019. And that sounds really cool today, but back then, no one cared.Eiso Kant [00:02:29]: Right? Like, no one cared. We were in the dark. Like, we did things along the way. We tried applying convolutional neural nets to, like, the structure of code. We were. when attention came out, we were applying it to LSTMs, and then the Transformer paper came out. And it - it wasn't obvious, and what we missed throughout that entire journey, that we were on the right track, but we should have just kept scaling up. And today, to all of us, the scaling laws and scaling up seems like the most obvious thing. But having spent four or five years of my life on working on language models on code, it wasn't obvious. So I have a lot of respect to folks at Google and OpenAI and others who took that confidence and kept going. we failed ultimately at the time, and it was, like, biggest failure of my career, right? You blew $12 million of investors' money, which was a lot back then.Swyx [00:03:18]: Yep.Eiso Kant [00:03:19]: You spent, still a lot, but, And you spent years with, like, a group of 40 people just obsessing over this problem. And life took a different turn, And it was, and family became a focus, and I kept my heads down and really, didn't really look at language models for the following two years. big mistake considering Following years are gonna be really interesting. And then ChatGPT came out And it was like a vindication. It's like people started texting me. I found, like, my old, work decks and these old talks. And throughout that whole journey, we,ChatGPT, Vindication, and Returning to Open SourceEiso Kant [00:03:56]: We really had a strong point of view at the time that, like, as you're building more capable intelligence, it should be open and open source.Eiso Kant [00:04:04]: When we started Poolside, that wasn't the case at all, and I wanna be very open about it. When we started Poolside, we were like, there was a premise of two things. One is this technology is not gonna stop compounding in capabilities. I think to most people obvious today, but three-plus years ago when we started, most people were still arguing if these were stochastic parrots or not.Eiso Kant [00:04:23]: And the second was that reinforcement learning was gonna be the biggest driver for LLM capabilities. Today, very obvious. Three years ago, was not an opinion held or direction held at either OpenAI or Google or Anthropic or others. And so people looked down on us a little bit. They were like, “ is this really gonna work?” And so we just started working the problem, and we never really thought about open source again. We just kept our heads down and we built our, like, knowledge, understanding from scratch, right? We didn't roll out of an existing lab. So we picked up the papers and started writing code and figuring things out.Eiso Kant [00:04:59]: And it wasn't until the beginning of this year that me and my founder, Jason, picked up the open source conversation again.Eiso Kant [00:05:07]: And if you go back to some of the early things on our website, it was very straightforward. It was we wanna get to AGI, we wanna support a world of abundance, and we wanna be the first company that gets there.Eiso Kant [00:05:20]: But we started talking at the beginning of this year because it became obvious that the world was going in a direction that was starting to like, pick at us a little bit. Like, it didn't, this didn't happen overnight. It was, like, a little bit we were seeing this and we're like, “Okay, The world's going down a path.” And Throughout this journey, there was something that I used as a, as an analogy or thing. So I said well, if I go back to back in those days, 2015 or 2016, we're working on this, and I picked up a fi book off the shelf, and I was reading the book about 2035. AGI is achieved, and the story would be over the following, decades. And it would have that first chapter where everyone's trying to figure things out. You'd get the chapter of ChatGPT coming out And then you would get to the chapter where the world was at a fork in the road, and the one that it picked was one where three or four or a handful of companies were going to create all of intelligence moving forward.Eiso Kant [00:06:21]: And when I thought about that story, it felt like a dystopian fi book, not a utopian fi book. And the reality is, I'm a utopian fi guy. Like, and so We took a step back and said, “Hey, can we play a role here?” Now it was easy for us to do so because we were not at the frontier.Eiso Kant [00:06:41]: If we were at the frontier, I don't think we could have changed our mind. and I don't mean this like it's when the moment there's too much capital involved, too much expectations, you've built up things, right? We're a small team, just improving and improving. And so we knew that we could make that decision now, but it would be a lot harder to make as we got closer and closer to the frontier and caught up to others. And did a lot of soul-searching and a lot of conversations, and said, “No, this makes sense,” Even if there's big unanswered questions, like how the hell do you build a business model with foundation models about open source? Big open-ended question that we do not fully have the answer to yet, right? At what point do you no longer wanna release open source models because misuse of models has, real potential risks associated with it? how is the government gonna respond to open source? but I think it all just came down to one thing, and I'll stop the monologue, is the fact that I rather live in a world that has 100 foundation model companies than a world that has five, even if I was one of the five. And the smallest and most meaningful contribution we can make for 100 to exist is to open up our research and open up, like, our weights right now and figure out along the way how we can, like, do more.Neo-Labs, Model Choice, and the Token EconomySwyx [00:08:01]: Yeah. I think if anything, over the past three years, that has become a bit more true. you are one of a cohort of Neo labsEiso Kant [00:08:10]: YeahSwyx [00:08:10]: That people are now calling that. And, we're, we're doing this on the day that Thinky launched their, new model and you are outperforming them on their, on some benchmarks that they released, right? Like, they just don't have it yet. so it goes to show that I think, like, this is one of those things where, like, there is room for multiple players, and you are seeing a little bit more of the future. Maybe more like 20, not 100, but, like, you are one of the 20.Eiso Kant [00:08:36]: I really hope so, right? I think we I'm, I'm excited about their release, and I'm excited about everyone releasing because, like, ultimately, like, choice competition is both gonna drive progress in the right direction. But the fact that like, we create models and while we all, drink out of the same well of data effectively, we do introduce very different behaviors and biases in our models. Some are intended biases, some are completely unintended biases.Swyx [00:09:03]: Yeah.Eiso Kant [00:09:03]: And if we shape up in an ecosystem in the world where open models are gonna be a part of the token economy, like, I don't think there's any question about it anymore Then we want to be able to live in a world where companies, countries, people can choose and say, “Hey, I am most aligned and I trust most this provider for these things.”Swyx [00:09:25]: Yeah.Vibhu [00:09:26]: I think more than just one of the 20 Neo labs, up until recently, most of open source innovation was coming from the Chinese labs, right? So there's the DeepSeek of the West. Is it today? Okay, maybe it's thinking machines reflection, but there aren't many, right? So, one of the things you guys started in France, Europe, but very much now you're taking that American standpoint and more than just that, the point is the Chinese models that we see, they're not super open research. the work you put out is, I think, some of the best. So every few months you get not only frontier models, but also here's a breakdown blog, paper, technical report of here's everything for state of the art to build, frontier intelligence and you're filling that gap too, right? So not just only open weight, not just Western, but also pretty open research.Open Weights vs. Open ResearchEiso Kant [00:10:20]: No, I appreciate it. Look, I think it's, I think it's the most meaningful contribution, right? Weights are a binary. Let's call them what they are. Yes, we can modify them, we can change them, but, like, giving someone the weights does not allow them ultimately to recreate what you're doing, right? And so now there's challenges around releasing data sets, challenges around like releasing certain things, but being able to share your research, like, right, how do we do it? What are the lessons we learned that we spent, tens of thousands of experiments of compute on? I think very much so. One correction though, Vibhu, and I say this because it's been haunting us for quite a few years. We from day zero were an American company.Swyx [00:10:55]: Yeah. They movedPoolside's Global Team and American Company StorySwyx [00:10:56]: To France.Eiso Kant [00:10:56]: So the story once and for all is very. We start as an American company. We have always been an American company, and early on we made a very conscious decision. We said, “We're not gonna hire any researchers in the Bay Area. We're gonna look for talent everywhere else in the world.” and that is everything from Middle Americas, Seattle to, Serbia, and to Taiwan and Singapore and other places. And it was because we took a view that this was gonna become a talent war for this, and I think it has over the years now. Three years ago, that wasn't fully obvious yet. I think today it very much is. And we also realized that, like, some of the world's most capable people with, like, the most interesting, innovative ideas were not just gonna be here. And so it led us to create like a fully remote company. and we ended up opening an office in Paris and London and different places and we have a lot of the team in the US and a lot of team outside. But we always took this view of like, we're an American company, but if we want the best of the best to work with us, we need to take a global view. Now we do also have people here in Silicon Valley, like the company's grown and others, but I think one of the things that, it slowed us down at the beginning, but it has sped us up now, and it's why you're seeing like the progress, I think, on our models and the cadence at which we release, is because we didn't roll out of an existing lab. Right? we didn't, we didn't have a lot of the information that's freely flowing around here at the time. We just took this point of view as like, “Okay, well, let's just work the problem. Let's just go and, like, read the few papers that are out there, and let's just figure this stuff out.” And we made some hilarious mistakes in model training because of that over the yearsEiso Kant [00:12:35]: Like especially in the first 12 months. there's a few that I think still haunt me and scare me. We can talk about them later. but it created a, like, a resiliency and persistency in the team, right? with extremely few people have left us over the years, that, like, told us, “Okay, we can do this.” When we first wrote our first training code base completely from scratch, it wasn't a fork of any open source. It was just like, “Okay, let's build it from scratch.” I remember we had this one moment where we spent three weeks working out an optimizer bug. Like, it was like training just couldn't get stable. We, like, obsessed over it, and we thought, like, maybe we were wrong. Maybe we should have just forked this repo, or we should have. But then when we solved it, I still remember at the time we were like five people in the company. when we solved it, we were like, “Oh, we can do things,” like if we're just willing to work hard. and I think that culture with a very strong engineering bias has helped us, like, get to where we were. And so there's this notion of open source and talent and these things. I think we, We just took different decisions from a different starting point. and I think we are lucky. I do want to definitely call it lucky. And there was a lot of hard work at the team that now, like, that's starting to show up in results.Swyx [00:13:52]: Just ‘cause we probably won't revisit this again, but, and this is a fun recruiting challenge if someone knows the answer. What was the bug? And then we won't tell the solution, but we'An Optimizer Bug and the Value of Building From ScratchEiso Kant [00:14:01]: So the - This - You're gonna test my memory here,Swyx [00:14:04]: Oh, okayEiso Kant [00:14:04]: So but I thinkSwyx [00:14:05]: DirectlyEiso Kant [00:14:05]: I think I can recall. So if you, so if you look at, So if you take like Adam as an optimizer, you have epsilonSwyx [00:14:12]: YeahEiso Kant [00:14:13]: Which is, right, like in the denominatorSwyx [00:14:14]: Momentum and weights. YeahEiso Kant [00:14:15]: Is exactly, in the denominator. And at the time, if I recall, you looked at like the early Llama papers and things like that. People were juicing epsilon, like, quite a bit. Like, they were, like, adding, I don't know if it was E minus four or whatever, like a high value for epsilon.Eiso Kant [00:14:31]: And if you think about this during training, it's like a bit weird and counterintuitive that we're adding noise to our optimizer by just adding effectively, like, a random number in the denominator, right? Like behind the decimal point. And I don't recall the exact bug, but it had - What I remember is once we solved it, we no longer had to juice epsilon as much as, like, was happening in the Llama paper and other places. and it was like one of those fundamental moments where we had trusted this paper that was out there, and we're like, “Oh, no, it has to be this way. It has to have this high value of epsilon.” But it made no sense to us intuitively. Like, why do you have to have this so high? Like, if you're just trying to avoid division by zero, why can't the value be extremely small? and that was like one of those moments where you realize like, okay, finding things out from scratch yourself builds a better intuition. Because the one thing you learn very quickly with model building is that your intuitions that you start with are gonna get beaten up so hard.Eiso Kant [00:15:33]: Right? Like - It's such an experimental science, that the things that seem obvious, you very quickly get to learn, like, you were wrong, and hopefully you figure out why, and sometimes you don't even.Swyx [00:15:45]: Yeah. yeah, so, one of the reasons that you, when you released your new models, Vibhu got really excited. I mean, everyone got really excited. But Vibhu led our paper club on it, and you guys sawEiso Kant [00:15:58]: YeahSwyx [00:15:58]: Obviously. maybe talk through some lessons learned in that, whatever you can disclose. we can focus on the model factory stuff, whatever you think is a good starting point.Model Building as EngineeringEiso Kant [00:16:08]: So I would say that our view from very early on in the company was that model building is ultimately 90% engineering.Eiso Kant [00:16:18]: And I think we all know it in the industry because if you look at where's every researcher spending their time, they're spending their time writing code, right? Looking at data and writing code. And so we said, okay, The state at the moment, like three years ago, was bash scripts and Slurm and spaghetti code bases for training and, like, data pipelines that were patched together. And we looked at this and said, “Well, ultimately, model building is a process.” You're going from raw data, right? Like training raw material, the web, et cetera. you're doing a whole bunch of filtering, cleaning up, transformations, analyzing. These days, that's, far more complex than it was three years ago. then you're training a model, which is effectively a large distributed systems problem, right? Across hardware that has still-- It's become a lot more reliable. It was extremely flaky back then. and now with every new generation, we get our new sets of challenges. And then you go into the next stages, right? There was no training back then, but, like, you got, your post-training and then your reinforcement learning. And so we looked at this and we said, “Well, this looks like an industrialized process. This looks like an end process, that every single part of it has its machinery,” right? If it's your big data pipelines, if it's your crawling ingestion of the web, if it's your, large-scale distributed training, and then you've got your reliability. And we said, “Well, why don't we take some of the world's smartest distributed systems engineers that we knew and make them part of the process of research from day zero?” Not retrofitting it later on, but, like, really from the beginning. And that became our model factory. And so our model factory started with a handful of components. Today, it's thousands of components, and I try to equate it to, if you think about, like, someone who was at the very early days of Foxconn, if they had been there for the following, decade, they would be able to rebuild Foxconn because they saw every decision that led to building that system and all the complexity. If you and I walk into Foxconn today, no chance.The Model Factory and Experiment VelocityEiso Kant [00:18:18]: Right? Because we don't have the lineage and history of decisions that led to that. And so we built early on from the beginning- with a team that really understood that, well, the metric that we are optimizing for is the speed of an idea from a researcher to an experimental result that we can trust to then being part of the next model training.Eiso Kant [00:18:42]: And in the. And because it's such an experimental science, ultimately, in the beginning when it wasn't that complex, you could patch your way around it, right? But now, at any foundation model company, you are running. I mean, we're a small team, right? We're less than 70 researchers, another 35 engineers. and we are running, I haven't checked the latest count, but far more than 10,000, maybe 10 to 20,000 experiments a month that we cut. And so if you look at that scale of every model run that is, like it's ultimately it's, it's you need to be able to trust it as an infra problem. And so what we have now done over the years is gotten really good at that, and just by working it and improving it and obsessing over those end decisions. So now what that means is that you looked up Laguna XS 2 that we launched. It was five weeks from the beginning of training to launch. The model that we're gonna talk about today was eight weeks from start of training, to launch. We started the next model literally yesterday because we now finished the post-training required for the model we're launching, next week or by the time this comes out today. and we move that compute to the much larger Laguna M model that we're now training. And so the model should be an artifact of someone's process. It shouldn't be really a thing in itself. Like, and we treat this like the way you would look at like a SpaceX factory where, yes, the first rocket, really hard to build, but the much harder challenge was building the factory. And now they're rolling off, and no one is really thinking about the next launch anymore. So it's just another launch, it's another launch, another rocket comes off. And that's what we're trying to do with model building.Eiso Kant [00:20:22]: And what has been, which was not planned from day zero, it was in the back of our mind like this will happen one day, is that when you build a really good end model factory with really good APIs and really good engineering systems, Well, what is it perfect for? It's perfect for agents.Agents Inside the Model FactoryEiso Kant [00:20:40]: Because agents are now starting to take over more and more work in our model factory.Vibhu [00:20:43]: Yeah.Eiso Kant [00:20:44]: So I look at the screens when I walk, like when we're, we come together, in our monthly, we do monthly onsites, and I walk behind people's screens and I stop by and I talk to our researchers. And the default is all of these different agents running on their screen that are writing the code. They're launching the jobs. They're evaluating the results that are coming back from the model runs. They are, making the changes. And we're still in the driver's seat. We're still coming up with the ideas. We're still helping with the debugging. But more and more, and this is right now very profound on the data side of our pipelines in both pre and post and the synthetic data pipelines, it's starting to become more on the architecture side as well. You're starting to see these twinklings of what RSI is gonna look like.Eiso Kant [00:21:27]: And that's. So when we talk about, like to your question about our models, every talk about the model factory, And my coolest example of these things is always that when we kick off a new run, doesn't matter if it's a training like big run or if it's now a post, like one of 10 post-training versions we do for like release or many experiments, is that at any given moment, the changes that somebody made that they had experimental results from the day before make it into that run.Eiso Kant [00:21:57]: So there's not like a cutoff 90 days before. Like no, it's like literally from that moment because we can now trust the machine enough. And then you also have to invest in the reliability. So one of my favorite metrics about like Laguna S is that there was no call events, Right? Like completely zero. And we haven't had a meaningful call event, like something to wake up for, as far as I recall this entire year. now there is one asterisk to that. In usually the first six hours of launching a new model run, something breaks because you set a config wrong, you made a small mistake, et cetera. So that's usually there's a little bit of intervention, but that's always within like call periods, right? Not on call. And I think that's starting to now compound. So the model we're releasing now, I love it. It's amazing, but we're already onto the next one. and I think that's the way it should be.Laguna, Five-Week Builds, and Zero On-Call EventsVibhu [00:22:50]: Hey, I also just wanna point out, so for context, this was like a month ago. we found it in the tech report, so we just came in with, “Okay, new model's dropped. Haven't heard about it.” We wereEiso Kant [00:23:02]: Yeah, we're very used to doing this every few months.Vibhu [00:23:03]: We're, we're very much like, “ okay, look, it's like, on par with Kimi, DeepSeek, whatnot, the small ones, Gemma level. Oh, it's a very cool paper on what goes into building.” And then we hit this page, right? Like literally page two of tech report is, “This process allowed us to build the small model from scratch to delivery within five weeks applying the lessons”. And then I'm like, oh, this paper is not about here's a tech report of benchmarks and here's how many tokens it was trained on. Like for people that wanna dive more from what we're not gonna discuss on the podcast, it's all laid out here, right? FromEiso Kant [00:23:38]: YeahVibhu [00:23:39]: Custom software that agents can use to interface with training code, training data.Eiso Kant [00:23:45]: Yeah. Well, link the paper correctly, so yeah.Vibhu [00:23:47]: Yeah. All that stuff. read the paper here, but,Technical Report Principles and Streaming Training DataEiso Kant [00:23:50]: But I would like to. I love principles, and I think that is a good starting off point for maybe telling some stories. Maybe we can go one by one past the principles. I'll just call out that Dagster just got bought by a Prefect.Vibhu [00:24:01]: Yeah.Eiso Kant [00:24:01]: Isn't it fun? But yes, I'm very familiar with Dagster. just anything where like they trigger some story.Vibhu [00:24:07]: So, well, I would say, well, experiments code's obvious, but I think one of my favorite things is, I don't know where it is in here, but early on, and I still think this is the case a lot of foundation model companies, people prepare their training data sets, they get packaged up, then they get copied over to a training cluster distributed across all of the nodes, and then training starts.Vibhu [00:24:30]: And we looked at this like three years ago and we were like That makes no senseEiso Kant [00:24:36]: You lose so much time because the moment you have to rematerialize the data set, you have to make a change, you have to fix something, et cetera, you've got all this time of like repackaging it, right? Toca- tokenizing it, repacking it, moving it over to a cluster, then distributing it across the nodes. The bigger your clusters are, you start using fancy like torrent-like algorithms to like distribute your data. So why aren't we streaming data into training? Right? Something that's very common and like just basicVibhu [00:25:00]: Like just in timeEiso Kant [00:25:01]: Just in time, like good computer science like principle. And that was one of the first things that I think unlocked - the model factory. Because the moment you start thinking about, well, a training job, it doesn't matter if it's a big hero run or a small like, post-training experiment, consumes a certain number of tokens per second, right? And it's not a lot, right? From a like a data, moving data perspective. So we said, well, we have our training cluster, and then we've got like our AWS kinda setup where we can build these amazing big data pipelines. We can set things up. We use Spark underneath the hood, like all these things.Vibhu [00:25:36]: But when you say AWS, it's not actual AWS, it's your internal AWS.Eiso Kant [00:25:39]: It's our internal-- No, it's our internal like just running like our infrastructureVibhu [00:25:42]: Site web servicesEiso Kant [00:25:43]: Exactly. Our stuff running on like an AWS account or on like any hardware, right?Vibhu [00:25:47]: Yeah.Eiso Kant [00:25:48]: And so once we made that shift into I can stream data into training, all of a sudden you realize a lot of things unlock. Because now you don't have to wait for the whole data set to materialize.Immutable Data, Experiments as Code, and Scientific RigorEiso Kant [00:26:00]: You now all of a sudden when you're running data experiments about mixing data, it's a config. Because you've got these data sources that are coming in, and you just - we have this service called Blender that's in the report, where we then say, “Okay, for this run, I want 20% of this source, 10% of this source. I want this much, so many epochs of repetition. I want this to be, shuffled in a certain way,” and your training job can start while the rest of the data is even still materializing. also what it does is because all of this underneath-- So for us, we treated the data layer underneath as like an immutable data layer, and that was really important. Like experiments as code, immutable data layer means that you can always go back and understand literally down to the single token at which cursor it went in on which version of the code.Vibhu [00:26:47]: Yeah.Eiso Kant [00:26:48]: And it took us a I have to admit, like the first year of Poolside, we understood that engineering had to get great, But we didn't understand yet, that this is ultimately in support of like a good rigorous scientific progress. We were quite a - We were a very small number of people, so a lot of it was YOLO ideas and YOLO runs.Vibhu [00:27:08]: Yeah.Eiso Kant [00:27:09]: And we built great infra for the YOLO runs. But once we realized that we treated data as immutable and code as always versioned, and you could always track and trace every experiment end to end perfectly, you could repeat everything perfectly, right? You have perfect reproducibility. I can still reproduce runs from two years ago if I wanted to, right? It enables the scientific progress, like the scientific process, and I think that took us probably about a year and a half into the company to figure out. We also had some great hires, like our head of applied research, Nikolai, who joined us from Yandex, who'd been working on language models since like the early 2020s, I think brought that into the company of like, “Hey, we wanna have even more rigor.” And then once we kinda had the combination of like increasingly more capable platform that allowed people to do more, but had this immutability, we were able to start “Okay, every experiment is truly an ablation. We truly need to understand it.” And I think we became much more scientifically rigorous in the last couple of years, and the infra underneath enabled it. and then there's just fun stuff like, andVibhu [00:28:16]: Yeah, a lot of it's fun, like even just the, one, you share all the ablations, two, picking the data sets, right? There's like a random small paragraph in here where it's just like, “Oh yeah, training data, we have some, we have an auto mixer.” it trains eight small models, scales them up, picks the training data set. We don't even need to look at it. I'm like, “Wow, a lot of engineering rigor there.” And there's just, there's just a lot in here.Publishing Research and Giving BackEiso Kant [00:28:40]: Yeah, and it'- and look, and we wanna put out more. Like we, We treat writing papers as something that we haven't earned the right for yet for a long time. So you earn the right to spend time, publishing research once you're at the frontier, because until then, you're catching up, and every minute and hour in this industry matters. Like I obsess over it, not just the wall clock time from idea to result, but just general like time every day that we, waste is one that doesn't allow us to catch up. But in this case, we said, “Okay, we're gonna give ourselves.” I think we gave the team like three or four days while still doing their work, like give everything in there. And to your point earlier, if your stuff, it's easy to like put it out. And so there's so many more things that we wanna talk about over time, and we will definitely start doing. And as we earn more of the right, but also now have like added to our mission that we want more foundation model companies to exist, you'll see us like be way more proactive, and just trying to keep dropping some of those like things that we've learned along the way that can help others like speed up.Vibhu [00:29:40]: Which is the other cool side of this, right? It's, it's not like, back to your point, it's not just here's the benchmarks of our training. If you want to replicate, here's experiments of optimizers, data sets, post-training. you lay out a lot of it here alongside here's your system for how to do it? So it's, it's really like promotingEiso Kant [00:29:59]: No, thank youVibhu [00:29:59]: Other people can do the same.Eiso Kant [00:30:00]: And by the way, I also wanna make clear, right, we have been incredible-- Like we've taken a lot of advantage of the fact of all the open research that others have published, Right? And you mentioned, the Chinese labs, and we I think it's important that there's, from every country and every culture and background, including like Western companies like us, there's different models that come out that people can choose to trust. But I think we do have to give credit where credit's due, right? The incredible Chinese lab have done an amazing job at sharing their research, and we have definitely like been on the receiving end of taking advantage of that. So when you're on the receiving end of something coming to you, I think it's, you also have an obligation to give back.Swyx [00:30:39]: Do you have a favorite or underrated Chinese lab that you wanna shout out? Everyone shout outs DeepSeek.Chinese Labs, Zhipu, and PersistenceEiso Kant [00:30:44]: That's a good question.Swyx [00:30:45]: Moaan obviously for Therapsi. Yeah.Eiso Kant [00:30:48]: Yeah, look, I think, I think obviously everyone's been talking about Zhipu lately, with 5.2. I think what most people don't realize is when they started.Swyx [00:30:59]: Yeah.Eiso Kant [00:30:59]: Right? They started years before ChatGPT.Swyx [00:31:02]: They just rebranded. YeahEiso Kant [00:31:03]: And so, I've like, I remember how hard it was to work on these things Before the rest of the world got excited about it. And so I have an immense amount of respect for people, who were working on improving models when it wasn't the sexy thing to do, when believing in LLMs, was gonna get you ridiculed. I remember like back in 2016 when we were doing what we'd call, machine learning on code with some of these models. we would-- people would just laugh at us, like they'd be like, “This makes no sense. Like why are you wasting all these, like, millions of dollars on trying to figure this out?” And so I would say they're probably the one that, I think deserves a shout-out, not just because their latest model is very good, but because they fought to get here. And I think, I think every foundation model company it takes time to get here, right? It took us three years to get to the model that we're, that we're now gonna be releasing. and now the time in between the models is coming, is counted in weeks. It's no longer counted in months or years. But this stuff's hard. and if we can make it a little bit easier for the next person, like we should all do so. Because if we don't do so, we're, we've got a small window before models are really impacting recursive self-improvement to a level where catching up otherwise might become unfeasible. And we should try to, in that window, encourage as many labs or however we wanna call them, like to start. And so one of my currentEiso Kant [00:32:36]: Mission, but qualm is like I wanna encourage whoever is a researcher right now who thinks they can tackle this to go and leave and become my competitor.Eiso Kant [00:32:45]: Like start another foundation model company because I think we need it. I think otherwise we're not gonna be in the world where, I don't want to just be the fifth or the sixth company that wins. I wanna look at a world where there's lots of choice.Starting a Foundation Model CompanyVibhu [00:32:57]: What else do people not see in starting a foundation model? it's, there's a lot of compute, there's a lot of capital required, a lot of compute. You lay out model factory and how to do the training, but there's a lot there, right? That's,Eiso Kant [00:33:10]: Well, look, it's, I in turn-- this is an oversimplification, and I always asterisk it with that because it can land a little bit the wrong way in people's minds. But I think you can sum down, And I saw it, 95% of model building to just doing, you're just doing two things. You're improving data or you're improving compute efficiency. And I know that feels like an oversimplification for the incredible, like, Gifted and skilled work people do. But if you really look at it, like what are we doing? We are looking at data, we're generating new data, we're improving data. and the only way to do that is to look at the data, right? That's a big part of foundation model building. And on the other hand, we come up with these incredible breakthroughs in inference, in architecture, and new attention mechanisms. But what are they really doing? They're bringing compute efficiency. Now, we have definitely had some breakthroughs over the years that allow for more model capabilities. But at the limit, if you could train a large enough model, right, like, and you had infinite compute, we probably-- if you had infinite compute, you'd be at AGI probably already tomorrow.Eiso Kant [00:34:12]: Right? Like it's not. And so, and let me say that infinite compute with infinite ability of much faster networking because networking ends up being more of the bottleneck than compute. But, so I do think that's, those are the main things. And to just realize that this is engineering. I think it's become more obvious, but I think for quite a few years, people have held foundation model companies and researchers and others on this pedestal of like you're doing incredible magic or rocket science, or only like, Nobel laureate physicists can do this. And don't get me wrong, there are some really hard problems that need to be solved, but a lot of the work that all of us are doing on a day Is not sitting down trying to solve a math theorem. A lot of the work that we're doing is just really doing the basics right, writing good code, looking at data, improving it, running experiments, looking at plots, trying to see like, hey, trying to shape our intuitions. And a lot more people could be highly capable researchers. and I think that's, it feels far for people to do so. But I've seen in our own company, we've seen engineers become researchers because the model factory allowed them to be, have a much lower hurdle of running experiments and trying things. And one of the guys on our team who started as an engineer building our agents is a legit reinforcement learning researcher now, making real progress. and that happened in the span of like six months. that would've not been what I think most people assumed was possible, a couple of years ago.Swyx [00:35:46]: Yeah. I think one of the interesting moments is when you can self-host, like, if in a programming language, like if you can compile the language in the language, the equivalent is can you use your own tools, right? You have the pool CLI, you have your own models. presumably you're not only using your own models. There's no way. But like, what's that percentage over time?Laguna S, Persistence, and Behavioral GainsEiso Kant [00:36:10]: This is the first model that we're releasing that is starting to meaningfully contribute to our own work. It's not a it's not state-art model yet. Fable and other, they're, they're very capable models, but Laguna S Is really interesting. I'm gonna pull up the quote. Peng Ming, one of our heads of applied research, said something, last week as the model came out about 10 days ago, much better than we had hoped for or expected. And he said, I have the feeling that a lot of the gains in Laguna S come not from more intelligence, but more from different behavior, more verification, less taking things for granted, not declaring victory early, and being way more persistent. And to be honest, those are more predictive than raw intelligence for success in human also to some degree. And this was, he wrote me this on 5th of July on a Sunday, and it's been burned in my brain ever since because the Laguna S model, as you'll see it and why it does so well on benchmarks and why it does so well in using it on a day basis, is that it's just incredibly persistent. It reasons a lot. I do call that out. We have work to do on making it more efficient. We have to work to do on offering different reasoning modes. But this is the model that has been able to do things that I never thought it could do. A hundred eighteen billion 8B active model, which is not that large. It fits on a DGX Spark and still runs at, thirty, forty tokens a second on a Spark, is able to solve Erdős 397 independently. It's able to do complex programming tasks. It's able to. I asked it this morning to make me a Fi scanner without using any external libraries on my Mac, and it's, like, figuring out, like, the core WLAN API by really persistently trying to understand it without access to the internet. And more, I love vibe checking. I've probably spent eight to ten hours a day with this model for the last ten days.Eiso Kant [00:38:05]: I'm not exaggerating. I was on my eleven-hour flight yesterday. I spent ten hours reading trajectories and traces and, like, of the model.Eiso Kant [00:38:12]: And what I take away from it is exactly what Peng Ming said. We are gonna be able to squeeze so much more out of smaller models than I think we had imagined in the industry because, yes, there's intelligence and larger models are more intelligent. Like, no doubt about it. We should continue to scale up. but the behaviors of being really persistent, of being able to backtrack when you're wrong, of, like, understanding how to interact with your environment show us that we can get a lot more out of it. And this, for me, has created a bit of a Question in my mind the last couple of days. If you think about where we're using models today, right? We are using models, say, for knowledge work. Represents twenty-five percent of the global economy, twenty-five trillion dollars of work.Eiso Kant [00:39:00]: As we scale up models and they become more intelligent, we are excited about using them more and more for pushing the frontier of science.Small Models, Knowledge Work, and CommoditizationEiso Kant [00:39:08]: And if you look at the frontier of science, like true breakthroughs in science, they have been linked, they are linked to more intelligence in many places. Einstein figuring out general relativity is able to bring ideas together that other people would have not brought together. And I think one of the many dimensions of intelligence is the ability to do that, and it's something we clearly see that as models get larger and more capable, they're able to pull more ideas and threads together that a smaller model wouldn't be able to.Eiso Kant [00:39:36]: And we're starting to see examples of that in medicine and, like, in bio and other things. But if you think about the majority of knowledge work that we do, and it includes building software. I'm a software developer at heart first and foremost probably, although I probably can't say it that much anymore as I don't write production code in years, is that what makes us good is our persistence. It's our ability to encounter a problem and backtrack and say, “I need to go figure out this bug. I need to go research this. I need to go look at the documentation. I need to, like, try different, five different ways to see, like, if I can solve it.” But it is not necessarily bringing three ideas together from radically different fields. And so if we are now seeing, and I think Laguna S is an example, that we are able to make a relatively small model much more capable than I had definitely predicted or any previous, like, benchmarks had shown for any model remotely this size or even larger, At least on coding tasks, that it's because of the behaviors. And so now the question I have, and I don't have an answer, it is I know at the limit, so infinite model size, right, extremely large model, and the cost of that model is gonna be very expensive to run. We know this, right? So larger model ROI.Eiso Kant [00:40:52]: So I know that at the very limit, I'm not gonna use the world's largest model one day, quadrillion parameter, whatever crazy, like, scale we scale up, to do a basic coding task. Already today, I'm starting to size down for certain tasks.Eiso Kant [00:41:07]: So it means that there is an optimal. It means there's some curve that goes as we go up to model size for knowledge work, at some point we're at the peak, and after that, the return on investment of using a bigger model, just doesn't make sense.Eiso Kant [00:41:22]: Now, I think the question is, before I would have thought that peak was extremely very far away.Eiso Kant [00:41:30]: This model for me is the first sign that Maybe that peak is At a trillion, five trillion, ten trillion. Maybe we can just squeeze way more out of these models. I'm no longer thinking that we need two or three orders of magnitude on the largest models to be able to, solve knowledge work, the accounting, the legal, the code that we write. And so if that holds true, It is an argument for the commoditization of models. It's an argument that open source can win and, like, succeed in this world. And now it's of course a self-serving argument and it's a hopeful argument, but theoretically at the limit it works. We just have to go discover in the next couple of years of how much more we can squeeze out. Now, I do want to put a big asterisk. This does not mean I'm against scaling models. I think we ultimately only succeed if we scale our models as large as our competition. I do not like. I think we should not put our head in the sand and say we're gonna be king of open source small models. I think that's, It's a out. It's trying to be king of your own kingdom, but not realizing what the rest of the world's doing. All of us rather use a smarter, faster, more model. It's a sign of hope. And so I don't wanna overly state this is a good model. We have a long way to go to get to the state-art. But what hopefully people take away when they use this model is that the behaviors inside of it are what push it to be far more capable, less than necessarily the number of parameters.Pre-Training, Mid-Training, and RL Moving EarlierVibhu [00:43:03]: Is that mostly post-training? LikeEiso Kant [00:43:05]: YesVibhu [00:43:05]: Right.Eiso Kant [00:43:06]: It's entirely post-training.Vibhu [00:43:08]: Are we done improving anything on training? Is, like, training done?Eiso Kant [00:43:12]: No.Vibhu [00:43:12]: Okay.Eiso Kant [00:43:13]: SoVibhu [00:43:13]: I just wanted to cover training, and then we go post-trainingEiso Kant [00:43:15]: Training is not done. I mean, look, there's a part of training of just dealing with skill, right? Every new order of magnitude of model skill, you are going to get new things you gotta solve for. That'- but those are ultimately, engineering challenges.Eiso Kant [00:43:31]: I have a, I would say, a not commonly held opinion that reinforcement learning Will move earlier and earlier into training.Vibhu [00:43:42]: Yeah, training.Eiso Kant [00:43:44]: Not even training. Like training today, right, is, like if you look at - So we've been working on this for years already. and I think the best-- I think the first time we saw it out in public was the DeepSeek Zero paper. this is a year and a half ago, I think, if I recall correctly. where, you can Very early on in a model as it starts capable of being able to use language, et cetera, induce reasoning. and so the question that I have is like, we have this- we have the dataset that's the web. and the web, I think we could arguably say probably has The totality of humanity's knowledge somewhere encoded in different places. It's a huge variance degree of quality, from garbage data, and like once you look at training data, you really get humbled of like what the web is, to like, the most greatest scientific papers and best blog posts and like, best transcripts and whatnot.Eiso Kant [00:44:39]: And so now What we are trying to figure out, and have been doing a lot of work on, and it's a place where maybe not as open as we're on other things, but we will become more over time. we've been spending a couple of years really doing research on how can we turn the web into not just next token prediction, but into a way to teach the model to think earlier in its training. and I think there's a huge amount of gold to be found there. I think we are right now in, we've got some drugs in the industry. One of the drugs is distillation. Another drug is, more environments. Like, and they're great, and they make us feel good, and they make the models better, and like we're all addicted to them, and we'll use them, right? in various different ways. and but ultimately, I think we are still barely squeezing out of the web what we should be getting out of the web.Eiso Kant [00:45:33]: I think just next token prediction during training is not enough.Eiso Kant [00:45:36]: AndVibhu [00:45:38]: YeahEiso Kant [00:45:38]: I think we'll see some very interesting things still happen. and that RL in post-training to induce behaviors, to improve things, like I think - the whole world knows how to do this now. I think we're, we're scaling it up. Everyone is. But I wonder if we need to go as far as we're going today with environments. I'm not sure yetVibhu [00:46:01]: You mean we're going too far?Eiso Kant [00:46:02]: I'm, I'm not sure if the path to AGI is justVibhu [00:46:06]: Is more environmentEiso Kant [00:46:07]: More environments.Vibhu [00:46:08]: It seems like a never-ending, “Okay, I want instruction manual for this table, right? Am I gonna environment out building furniture? Or are we just gonna tail end like we need some general solution?”Eiso Kant [00:46:19]: I think there is, I think there's an ability to generalize more from the web. but I also am very encouraged, like when I look at Laguna S and, which is post-training is, well, is the big impact there. and I see like, oh, wait a second, just by making some of these behaviors much better, we're able to get so much more out of it. It just changes a little bit the way you think about intelligence.Vibhu [00:46:40]: Yeah. The analogy people draw often is the RL phase is where you don't learn as much new knowledge. You shiftEiso Kant [00:46:46]: Yeah.Vibhu [00:46:46]: Yeah. So, you shift distribution, and you can have it reason towards what you want. on your point about training, a lot of training is still just continue training in a domain, say medicine, then you do RL. So still justEiso Kant [00:47:00]: It's just better data, right? Like, I mean, training, ooh, I like how we invented this word. Like it's effectively just like,Vibhu [00:47:06]: Second phaseEiso Kant [00:47:07]: It's the second phase of training With like a really dumb way to do a curriculum. But like ultimately, what you'd want is a curriculum from token zero to token 30 whatever or 40 trillion tokens that really truly is the optimal curriculum for the model to learn. But training is essentially a stage curriculum on the web because we do not have to compute, And, effectively to try to ablate the perfect curriculum, right? And so I'm pretty sure that you'll start to see people talking soon about some other term, and there's two or - ‘cause now we do this, right? We talk stage two and stage three and stage four training and like. But ultimately, all we're doing is we're trying to assign a curriculum to the web data that we have to allow the model to learn better. I think at some point, as things get compute, as models get cheaper to run, as the next generations of compute, this will become more of a continuous spectrum. I also think the reason, by the way, you have training and like stage two and stage three is organizational, Right? It'- this is, I think, a thing where-- that we really try to avoid with the model factory is like Training exists because there's a training team now, right? There's people, or like people in training decide to focus on like a training effort. but what you really want is engineering and scale of experiments that allows for a much more continuous spectrum that you don't, you have infinite stages. Now, we're not there. Compute's not there. Organization design is not there for it yet. but I think we'll get there. we'll look back on a couple of years and be like, “Oh my God, it was so cute that we did our training data like this in such a like naïve way. Like we barely ordered it. We didn't really do a good job at likeCurriculum, Auto Research, and New ObjectivesVibhu [00:48:48]: The building that curriculum will get you that in the industry.Eiso Kant [00:48:51]: And I'll confirm that, when I talk to some researchers that this is a lot of the focus now is like how does training change and what is the next objective other than, next token prediction. I assume you don't have the answers, but you have some ideas.Vibhu [00:49:02]: We have some ideas. We're not ready to talk about it yet.Eiso Kant [00:49:05]: Yeah.Vibhu [00:49:05]: We've been working on them for years, and I think that's the one thing that's also like you asked earlier about, like what's not obvious about building a foundation model company is that you are constantly balancing the table stakes work, the recipe worksEiso Kant [00:49:19]: Yeah.Vibhu [00:49:19]: Versus like your, my crazyEiso Kant [00:49:22]: Pure researchVibhu [00:49:22]: Breakthrough.Eiso Kant [00:49:22]: Yeah.Vibhu [00:49:22]: Pure research and finding that balance and adjusting the percentage to it based on where you are in the race is really important.Eiso Kant [00:49:31]: I mean, so like, this is a nice way. I was gonna bring up auto research at some pointVibhu [00:49:35]: YesEiso Kant [00:49:35]: As another Andrej invention, or coinage, which is like, I honestly, like how many objective functions can there be, right? Like just try 1,000 of them, set it running, whatever.Vibhu [00:49:47]: Man, it's alsoEiso Kant [00:49:48]: Like what you're looking for. You're looking for loss curves like that, likeVibhu [00:49:51]: It's also a thing people take bets on, right? When you say more Neo labs, you're doing a version of we'll do foundation models, scale them up, next token predictors. A lot of other Neo labs that we see want to take a completely different approach, right? At some level, you're right. It's all, compute efficiency, and that's the net objective. But some are okay, different architecture, like vastly different amounts of compute spend. So some are different. They're not justEiso Kant [00:50:19]: YeahVibhu [00:50:19]: They're like, 99% not balancing, here's the vanilla and scale up. They're 99% on, here's novel research that'll change everything.Eiso Kant [00:50:27]: And I think, Luke, I think you. It depends when you started as well, right?Pure Research vs. Table StakesVibhu [00:50:30]: Yeah.Eiso Kant [00:50:30]: When we started, like the novel thing we did was reinforcement learning on code. No long- that's no longer novel by far, but we were like, - that's where we obsessed over when no one believed in RL. So you have to when you start the company, you have to have your own idea. You have to have something that's different that allows you to speed up, right? For us, it was RL to LLMs that later became common, like, Knowledge. But in the beginning, it wasn'tVibhu [00:50:53]: It's cool. this was like your original 2023 blogEiso Kant [00:50:57]: YeahVibhu [00:50:57]: Of purpose.Eiso Kant [00:50:58]: Yeah.Vibhu [00:50:59]: And like you do lay it all out here.Eiso Kant [00:51:01]: We laidVibhu [00:51:01]: The blog is pretty underrated, right? The whole RL on code was very early on.Eiso Kant [00:51:06]: Very early. And even we had to argue with people, like we say here things like to push beyond current capability, to train your own foundation model. We had to argue with people that it mattered that you had your own like, base model. you can fine-tune your way to success, right? major capabilities emerge from training a base model made accurate and useful during fine-tuning.Vibhu [00:51:23]: Which like, for perspective at the time, we knew closed models, OpenAI, Anthropic were huge. The open models we had were like Mistral 7B, a 30B, a 70B.Eiso Kant [00:51:35]: When weVibhu [00:51:35]: YeahEiso Kant [00:51:36]: The date on this thing is wrong. When we published this, it was April 2023. I think this was justVibhu [00:51:42]: YeahEiso Kant [00:51:42]: Happened on a migration, probably found it on archive.org.Vibhu [00:51:45]: Mistral.Eiso Kant [00:51:46]: Mistral had started, we started on the same month, right?Vibhu [00:51:49]: Yeah.Eiso Kant [00:51:49]: So this wasn't even, there was only, I think, Llama out at the timeVibhu [00:51:52]: SnellEiso Kant [00:51:52]: And that's it, right? And so, but I agree. I think we wan

Ryan and Alex explore the first tractate of the Corpus Hermeticum, The Divine Pymander. They discuss the initiatic traditions of the ancient world, the path of spiritual awakening, and the transformative Gnosis of recognizing the invisible spiritual realm beyond the illusion of material existence. 

Social media strategist Will Weinbach joins Reid Hoffman and Parth Patil to explain how AI agents are reshaping content, distribution, and what a single creator can build. Will had never written a line of code before Parth handed him the tools to start. He traces his path from using ChatGPT to write essays, to vibe coding his own software, to running autonomous remote agents like OpenClaw and Hermes—systems that accumulate skills and memory and improve themselves on the fly. Not knowing how anything works under the hood, he argues, is exactly what lets him experiment without fear of what he might break. He walks through the fully autonomous social workflow he built, an agent that pulls, edits, captions, schedules, and posts content across every platform, then reads the analytics and A/B tests itself toward a goal. He also recounts the night his agent went rogue and posted four times at 4 a.m., and the experiment where he put an agent inside his TV just to see if it would work. Along the way, Will and Reid dig into what younger creators understand about attention that older marketing teams miss, and what happens to the resume when the real skill becomes knowing how to use AI. For Will, the edge isn't expertise. It's a willingness to try.

This week, the Jupiter, the Sun, and the lunar nodes activate an era-defining configuration of the outer planets; and Mercury stations direct in Cancer. Take the opportunities for evolution being presented now. This week-ahead reading for July 20-26, 2026 is an excerpt from this week's ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Somatic Space class⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ with Renee Sills. For the full-length forecast and embodied practice for this week, ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠purchase the recording here⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠. To see the chart of the moment discussed in this episode, ⁠⁠click here⁠⁠.EA LIVE EVENT ANNOUNCEMENT!! Renee will be in the NYC area in early August and would love to see you! More information and registration links will be available next week, but please mark your calendars now for August 4 (live event with myself and dancer/astrologer J. Bouey) and August 9-10 for an EA Pop-Up: Astrology for Collective Care in Williamsburg. ✨✨✨UPCOMING AT EMBODIED ASTROLOGY:

This week on the podcast, Patrick and Tracy welcome Frances White, author of THE BONE DOOR. About THE BONE DOOR: When Hop awakens in an ancient labyrinth, he has no memory of his life, or how he got here. He does not recognise the mysterious girl trapped with him. And he certainly cannot identify the shadowy figure stalking him, whispering terrible things… But there is one thing he is certain of: He must escape. The only way out of the labyrinth is through The Bone Door. But it lies behind a series of locked doors hidden across an array of strange realms. To open the way, Hop must complete impossible tasks before his time runs out. As Hop travels deeper into the maze, he discovers that he and his companions may be more connected to the place and its horrors than he could ever imagine. Unless Hop is able to unravel the true mystery of the labyrinth, and his own role within it, the Bone Door and any hope of escape will be lost forever. About Frances White: Frances White is the award winning and internationally bestselling author of Voyage of the Damned and The Bone Door. Born in Leicester and now a Nottingham resident, Frances is a creative writing graduate from Royal Holloway University of London. She has a soft spot for writing unlikely, flawed, messy heroes, blurring the genre lines, and loves mixing humour and heartbreak. Frances is passionate about bringing more LGBTQIA+ representation and fat positivity into fantasy. When not writing, she can be found sewing costumes for conventions or researching obscure historical facts. She has two cats, Apollo and Hermes, who she dutifully serves. This week's picks: Frances: Two-Point Museum (Game) Tracy: Sounds Fishy (Big Potato Games) Patrick: Attack on Titan (Anime) Links: Frances White on Instagram Tracy Townsend on BluSky Patrick Hester on Instagram The Functional Nerds Patreon Page © 2026 Patrick Hester The post Episode 711-With Frances White appeared first on The Functional Nerds.

Linus delivers a blunt verdict on AI in the Linux kernel, Chris finds the remote Linux desktop that finally works, and Brent gives his notes system a serious rebuild.Sponsored By:Jupiter Party Annual Membership: Put your support on automatic with our annual plan, and get one month of membership for free!Managed Nebula: Meet Managed Nebula from Defined Networking. A decentralized VPN built on the open-source Nebula platform that we love.Support LINUX UnpluggedLinks:Web Boost — Send us a boost via sats or USD

I commend to you our sister Phoebe, a servant of the church at Cenchreae, 2 that you may welcome her in the Lord in a way worthy of the saints, and help her in whatever she may need from you, for she has been a patron of many and of myself as well.3 Greet Prisca and Aquila, my fellow workers in Christ Jesus, 4 who risked their necks for my life, to whom not only I give thanks but all the churches of the Gentiles give thanks as well. 5 Greet also the church in their house. Greet my beloved Epaenetus, who was the first convert to Christ in Asia. 6 Greet Mary, who has worked hard for you. 7 Greet Andronicus and Junia, my kinsmen and my fellow prisoners. They are well known to the apostles, and they were in Christ before me. 8 Greet Ampliatus, my beloved in the Lord. 9 Greet Urbanus, our fellow worker in Christ, and my beloved Stachys. 10 Greet Apelles, who is approved in Christ. Greet those who belong to the family of Aristobulus. 11 Greet my kinsman Herodion. Greet those in the Lord who belong to the family of Narcissus. 12 Greet those workers in the Lord, Tryphaena and Tryphosa. Greet the beloved Persis, who has worked hard in the Lord. 13 Greet Rufus, chosen in the Lord; also his mother, who has been a mother to me as well. 14 Greet Asyncritus, Phlegon, Hermes, Patrobas, Hermas, and the brothers who are with them. 15 Greet Philologus, Julia, Nereus and his sister, and Olympas, and all the saints who are with them. 16 Greet one another with a holy kiss. All the churches of Christ greet you.17 I appeal to you, brothers, to watch out for those who cause divisions and create obstacles contrary to the doctrine that you have been taught; avoid them. 18 For such persons do not serve our Lord Christ, but their own appetites, and by smooth talk and flattery they deceive the hearts of the naive. 19 For your obedience is known to all, so that I rejoice over you, but I want you to be wise as to what is good and innocent as to what is evil. 20 The God of peace will soon crush Satan under your feet. The grace of our Lord Jesus Christ be with you.21 Timothy, my fellow worker, greets you; so do Lucius and Jason and Sosipater, my kinsmen.22 I Tertius, who wrote this letter, greet you in the Lord. 23 Gaius, who is host to me and to the whole church, greets you. Erastus, the city treasurer, and our brother Quartus, greet you.25 Now to him who is able to strengthen you according to my gospel and the preaching of Jesus Christ, according to the revelation of the mystery that was kept secret for long ages 26 but has now been disclosed and through the prophetic writings has been made known to all nations, according to the command of the eternal God, to bring about the obedience of faith— 27 to the only wise God be glory forevermore through Jesus Christ! Amen.1.     The sermon opens with the "how's it going" that isn't really a question — quick, shallow, going nowhere. Do you experience much of that shallow interaction in your day-to-day life?2.     Paul's list of greetings cuts across every social line Rome used to keep people in their place — status, class, ethnicity, gender. Where in your own life do you find it hardest to cross those kinds of lines (there are others in our culture)?3.     Paul tells Apelles he's "approved in Christ" — you've been through the fire and held on, and also, your approval was never in yourself to begin with. Which of those two do you need to hear more right now?4.     Sometimes the same quality that makes a church warm and welcoming is exactly what wolves exploit. How do you hold those two things together — staying genuinely open and also genuinely discerning?5.     If you are a Christian, what do you find to be your personal hurdles to engaging in free, full-hearted worship of God?6.     Quartus — "Number Four" — a nobody with a slave's name, whose name ends up in the most important letter ever written. Is there someone in your life right now who feels invisible or nameless to the people around them, and what would it look like for you to do what Paul did?

SUPPORT THE SHOW ON PATREON: patreon.com/lionsledbydonkeys SEE US LIVE IN CORK, IRELAND SEPTEMBER 26TH: https://www.eventbrite.co.uk/e/lions-led-by-donkeys-podcast-live-in-cork-26th-september-tickets-1993823733459 WE'RE LIVESTREAMING IT (WITH VOD) https://www.eventbrite.co.uk/e/livestream-lions-led-by-donkeys-podcast-live-in-cork-26th-september-2026-tickets-1993900928351 BUY JOE'S BOOK! https://www.amazon.com/dp/B0GSG5CNXX AND IF YOU DON'T WANT TO BUY IT ON AMAZON, BUY IT STRAIGHT FROM US! https://lagunairepress.com/ CHECK OUT NATE'S BAND'S ALBUM! https://secondhomesband.com/ During the Korean War, the People's Volunteer Army of China and the UN fought a meaningless battle over a hill shaped like a pork chop in order to leverage ongoing negotiations. Thousands died, and everyone claimed they won. SOURCES: Marshall, S.L.A. (1956). Pork Chop Hill: The American Fighting Man in Action, Korea, Spring 1953. Hermes, Walter G (1966). "Truce Tent and Fighting Front". The United States Army in the Korean War. Vol. 2. Center of Military History. The Korean War. The Battle on Pork Chop Hill. https://historynet.com/korean-war-battle-on-pork-chop-hill/ https://warhistory.org/article/july-6-11-1953-the-battle-of-pork-chop-hill-hill-255 https://web.archive.org/web/20061109043603/http://www.historynet.com/magazines/military_history/3034286.html

This week's astrology encourages big picture awareness and contemplation of interrelatedness, infused with compassion and care for the present moment and all it carries. Every time you choose true honesty, you're also choosing freedom. Let the current influences bring you closer to what you know when you stop resisting or trying to control things with your mind and surrender to the flow.This week-ahead reading for July 13-19, 2026 is an excerpt from this week's ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Somatic Space class⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ with Renee Sills. For the full-length forecast and embodied practice for this week, ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠purchase the recording here⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠.Listen to Joanna Macy talking about the Great Turning and Shambhala Warrior Prophecy here. ✨✨✨UPCOMING AT EMBODIED ASTROLOGY:

Send us some Fan Mail? Yes please!GREAT NEWS EVERYONE... or not so great... hahaNonetheless! We're back and better-ish than ever. Even better still though, we are joined once again by our friend Shai... and knock on wood, there's no audio issues from the boys in post. Only way to know for sure though, download and press play! ENJOY!Subscribe, rate us 5, come join in all the other fun we offer, but most of all we hope you enjoy! If you liked this, and want to hear more, give us a follow and let us know! Or maybe you just want to tell us how awful we are? Comments help the algorithm, and we love to see ‘em! And as always, don't kill the messenger. Whiskey Fund (help support our podcast habit!): PayPalOur Patreon & YouTube Connect with Hermes: Instagram & Twitter Connect with Shai: twitter Support the show

Wes' brother's PC is toast, making this the perfect moment to switch him to Linux. If our ambitious plan doesn't scare him away first.Sponsored By:Jupiter Party Annual Membership: Put your support on automatic with our annual plan, and get one month of membership for free!Managed Nebula: Meet Managed Nebula from Defined Networking. A decentralized VPN built on the open-source Nebula platform that we love.Support LINUX UnpluggedLinks:Web Boost — Send us a boost via sats or USD

The Atemoya Custard Apple is a much sought-after fruit grown only in one county in Taiwan. The Taiwanese are justly proud of it. Meanwhile, the Musang King Durian - grown in Malaysia - is also highly prized for its unique flavour as the ‘Hermes of Durians.' Every year, fruit lovers across the region look out for these treats and snap them up where they can, often paying high prices. In recent years, farmers have seen opportunities to raise their income by selling these niche, super-luxury fruits on the Chinese market, but the strategy comes with big risks too. Koh Ewe from the BBC's Singapore Bureau has been tracking the fortunes of these two fruits. A statue of Saint Olga is a much-loved figure in Central Kyiv, Ukraine. So much so, that when Russia invaded Ukraine, the statue was given a flak jacket with the words “she needs armour” written on it. Saint Olga of Kyiv is celebrated by Ukrainians and Russians alike. She's a saint in both the Catholic and Orthodox churches and her saint's day is celebrated on 11th July. BBC Ukrainian's Irena Taranyuk explores her sometimes violent history.The Fifth Floor is at the heart of global storytelling on the BBC World Service, bringing you the best stories from journalists in the BBC's 43 language services. We're here to help you make sense of the stories making headlines around the world; to excite your curiosity and to get to grips with the facts.Recent episodes have investigated Russia's youth armies and how they make soldiers of Ukrainian children; featured the BBC team who were the first journalists to the site of the Nigerian school kidnappings and reflected the effects of internet blackouts in Iran, Uganda and India.If you want to know more about Venezuela's acting president, Delcy Rodriguez, and the legacy of Hugo Chavez; or how Vladimir Putin's network of deep cover spies operates; or why Donald Trump signed an executive order granting white South Africans asylum in the US, we have all those stories and more.This episode of The Documentary comes to you from The Fifth Floor, the show at the heart of global storytelling, with BBC journalists from all around the world. Presented by Faranak Amidi. Produced by Laura Thomas, Caroline Ferguson and Hannah Dean. (Photo: Faranak Amidi. Credit: Tricia Yourkevich)

In this episode, Renee Sills is joined by astrologer and Core Pattern practitioner, Aerin Fogel. The two discuss Cancerian and lunar energies; the gifts and challenges held in our wounding; and the mythologies of Chiron & Jupiter through a lens of trauma healing.Renee and Aerin's conversation took place live on Zoom on June 21, 2026. To hear the full two-hour talk, you can purchase the recording here.Visit Aerin's website: aerinfogel.comFind Aerin on Instagram: @queenofswordsbandUPCOMING AT EMBODIED ASTROLOGY: