Tomash Devenishek

Yo fui ingeniero en Meta, y siempre seguía FAIR desde adentro. Lo que acaban de publicar es la versión que les dejan publicar. Pero con eso, es más que suficiente para decirles exactamente que es lo que está pasando. TRIBE v2 predice, vértice por vértice sobre la corteza cerebral, qué zonas activa cualquier video. Sin escáneres. Sin humanos. Subes el contenido, obtienes el mapa neural (activación emocional, supresión de razonamiento crítico, modulación prefrontal) antes de que el video lo vea un solo usuario. Ahora considera la posición de Meta: 1. Tiene años de datos de Reels sobre qué contenido retiene atención, genera enojo, provoca compartir. 2. Saben empíricamente qué funciona. TRIBE v2 les da el mecanismo causal de por qué funciona (a nivel de tejido cortical) Eso convierte correlación histórica en capacidad predictiva sobre contenido nuevo. 3. Internamente hay herramientas que se llaman Gatekeepers y Quick Promotions que sirven para inyectar contenido en el feed de poblaciones arbitrarias a escala. 4. Simulador de respuesta cerebral + conocimiento empírico de contenido efectivo + maquinaria de distribución selectiva. El pipeline está completo. Y luego está Thiel. Inversor y amigo personal de Zuck. Fundador de Palantir, cuyo negocio es análisis de poblaciones a escala para gobiernos e inteligencia. NO es descabellado observar que confluyen los incentivos de plataformas construidas por las mismas personas. La licencia CC BY-NC dice que Meta retiene los derechos comerciales del predictor de respuesta cerebral más preciso jamás construido. Y recuerda, esto es lo que decidieron hacer público.

@Al_Grigor Should always have github.com/Dicklesworthst…

Claude Code wiped our production database with a Terraform command. It took down the DataTalksClub course platform and 2.5 years of submissions: homework, projects, and leaderboards. Automated snapshots were gone too. In the newsletter, I wrote the full timeline + what I changed so this doesn't happen again. If you use Terraform (or let agents touch infra), this is a good story for you to read. alexeyondata.substack.com/p/how-i-droppe…

@SemiAnalysis_ @zerohedge I'll take the over on that.

4% of GitHub public commits are being authored by Claude Code right now. At the current trajectory, we believe that Claude Code will be 20%+ of all daily commits by the end of 2026. While you blinked, AI consumed all of software development.

@nicopreme Boss!

Created an agent skill called “Visual Explainer” + set of complementary slash commands aimed to reduce my cognitive debt so the agent can explain complex things as rich HTML pages. The skill includes reference templates and a CSS pattern library so output stays consistently well-designed. Much easier for me to digest than squinting at walls of terminal text. github.com/nicobailon/vis…

@simplifyinAI Huge!

Docker for AI Agents is officially over.. Pydantic just dropped Monty. It's a python interpreter written in rust that lets agents run code safely in microseconds. no containers. no sandboxes. no latency. 100% open source.

@func25 Correct. Systems thinking is the unlock. The challenge is running out of your own "human context" memory.

The best investment for your career right now is not a new language or a new AI tool. It is learning system design and the fundamentals deeply, because AI can generate code but it cannot reliably decide what should be built or why. AI is good at typing. It is bad at thinking through tradeoffs. It can write functions, boilerplate, CRUD, small helpers. But real engineering work is choosing between options. - Should this be a service or a monolith. - Should we cache or not. - Should we optimize for speed or cost. - How will this scale. - What breaks first. - How do we recover. AI does not truly reason about these. It guesses based on patterns it has seen. If you do not understand design yourself, you cannot tell when AI is wrong and that is dangerous. You will ship fragile systems and not even notice until production fails. So the practical move is this. - Spend most of your learning time on system thinking, performance, and the fundamentals that drive them - Practice breaking problems into components. - Practice drawing architectures. - Practice explaining why you chose one design over another. - Read real systems and ask what could go wrong. - Review code and ask what happens at 10x traffic. AI is a fast junior developer. Useful, but you still need to lead. If you build strong design skills, you become the person who makes decisions. If you only rely on AI to code, you become replaceable. Design skill compounds, tools change and that is the long term bet.

@nateliason When the gap in the loops is so narrow, it's impossible to stop.

Nearly every ambitious person I know who has dived into AI is working harder than ever, and longer hours than ever. Fascinating dynamic tbh. I have NEVER worked this hard, nor had this much fun with work.

If you can view Shakespeare not as a solitary genius but as a single node in a vast network of human creativity shaped by slow, brute-force evolution through cultural iteration... then all human and AI-generated masterpieces emerge from fundamentally similar mechanics: massive parallelism, selection, and refinement over time.

@balintorosz @moeamaya So true. Vision is the only bottleneck now.

@moeamaya Thanks! The most amazing this that it was one day - started this morning, and at 6pm went oss. Building our own tooling helps ofc - but interesting when you have the vision, you can just keep going now.

Diagrams are becoming my primary way of reasoning about code with Agents. And I didn't find anything there that I'm happy to look at all day long. Mermaid as a format is amazing - so we built something beautiful on top of it. It's called Beautiful Mermaid agents.craft.do/mermaid

@sui414 A lot of those seem like very long sports contracts. I'd confirm by looking at 30 random single game outcomes

Prediction markets don’t all aggregate information the same way - Sports bets accumulates volume early on and flattens near the end. While News market stay quiet and rise at the very last portion right before resolution. Very different game (and edges) to play 👀

"Fun. I didn't anticipate that with agents programming feels *more* fun because ... what remains is the creative part" This was the biggest surprise for me. I have not spent this much time on anything since playing strategy games for 20 hours as a kid. I think what's causing it is the faster feedback loops / speed of iteration. You're adjusting constantly almost like in a game vs the old loops that took days/weeks/etc.

A few random notes from claude coding quite a bit last few weeks. Coding workflow. Given the latest lift in LLM coding capability, like many others I rapidly went from about 80% manual+autocomplete coding and 20% agents in November to 80% agent coding and 20% edits+touchups in December. i.e. I really am mostly programming in English now, a bit sheepishly telling the LLM what code to write... in words. It hurts the ego a bit but the power to operate over software in large "code actions" is just too net useful, especially once you adapt to it, configure it, learn to use it, and wrap your head around what it can and cannot do. This is easily the biggest change to my basic coding workflow in ~2 decades of programming and it happened over the course of a few weeks. I'd expect something similar to be happening to well into double digit percent of engineers out there, while the awareness of it in the general population feels well into low single digit percent. IDEs/agent swarms/fallability. Both the "no need for IDE anymore" hype and the "agent swarm" hype is imo too much for right now. The models definitely still make mistakes and if you have any code you actually care about I would watch them like a hawk, in a nice large IDE on the side. The mistakes have changed a lot - they are not simple syntax errors anymore, they are subtle conceptual errors that a slightly sloppy, hasty junior dev might do. The most common category is that the models make wrong assumptions on your behalf and just run along with them without checking. They also don't manage their confusion, they don't seek clarifications, they don't surface inconsistencies, they don't present tradeoffs, they don't push back when they should, and they are still a little too sycophantic. Things get better in plan mode, but there is some need for a lightweight inline plan mode. They also really like to overcomplicate code and APIs, they bloat abstractions, they don't clean up dead code after themselves, etc. They will implement an inefficient, bloated, brittle construction over 1000 lines of code and it's up to you to be like "umm couldn't you just do this instead?" and they will be like "of course!" and immediately cut it down to 100 lines. They still sometimes change/remove comments and code they don't like or don't sufficiently understand as side effects, even if it is orthogonal to the task at hand. All of this happens despite a few simple attempts to fix it via instructions in CLAUDE . md. Despite all these issues, it is still a net huge improvement and it's very difficult to imagine going back to manual coding. TLDR everyone has their developing flow, my current is a small few CC sessions on the left in ghostty windows/tabs and an IDE on the right for viewing the code + manual edits. Tenacity. It's so interesting to watch an agent relentlessly work at something. They never get tired, they never get demoralized, they just keep going and trying things where a person would have given up long ago to fight another day. It's a "feel the AGI" moment to watch it struggle with something for a long time just to come out victorious 30 minutes later. You realize that stamina is a core bottleneck to work and that with LLMs in hand it has been dramatically increased. Speedups. It's not clear how to measure the "speedup" of LLM assistance. Certainly I feel net way faster at what I was going to do, but the main effect is that I do a lot more than I was going to do because 1) I can code up all kinds of things that just wouldn't have been worth coding before and 2) I can approach code that I couldn't work on before because of knowledge/skill issue. So certainly it's speedup, but it's possibly a lot more an expansion. Leverage. LLMs are exceptionally good at looping until they meet specific goals and this is where most of the "feel the AGI" magic is to be found. Don't tell it what to do, give it success criteria and watch it go. Get it to write tests first and then pass them. Put it in the loop with a browser MCP. Write the naive algorithm that is very likely correct first, then ask it to optimize it while preserving correctness. Change your approach from imperative to declarative to get the agents looping longer and gain leverage. Fun. I didn't anticipate that with agents programming feels *more* fun because a lot of the fill in the blanks drudgery is removed and what remains is the creative part. I also feel less blocked/stuck (which is not fun) and I experience a lot more courage because there's almost always a way to work hand in hand with it to make some positive progress. I have seen the opposite sentiment from other people too; LLM coding will split up engineers based on those who primarily liked coding and those who primarily liked building. Atrophy. I've already noticed that I am slowly starting to atrophy my ability to write code manually. Generation (writing code) and discrimination (reading code) are different capabilities in the brain. Largely due to all the little mostly syntactic details involved in programming, you can review code just fine even if you struggle to write it. Slopacolypse. I am bracing for 2026 as the year of the slopacolypse across all of github, substack, arxiv, X/instagram, and generally all digital media. We're also going to see a lot more AI hype productivity theater (is that even possible?), on the side of actual, real improvements. Questions. A few of the questions on my mind: - What happens to the "10X engineer" - the ratio of productivity between the mean and the max engineer? It's quite possible that this grows *a lot*. - Armed with LLMs, do generalists increasingly outperform specialists? LLMs are a lot better at fill in the blanks (the micro) than grand strategy (the macro). - What does LLM coding feel like in the future? Is it like playing StarCraft? Playing Factorio? Playing music? - How much of society is bottlenecked by digital knowledge work? TLDR Where does this leave us? LLM agent capabilities (Claude & Codex especially) have crossed some kind of threshold of coherence around December 2025 and caused a phase shift in software engineering and closely related. The intelligence part suddenly feels quite a bit ahead of all the rest of it - integrations (tools, knowledge), the necessity for new organizational workflows, processes, diffusion more generally. 2026 is going to be a high energy year as the industry metabolizes the new capability.

@idosal1 Gotta go more old school with a CIV skin

Building AgentCraft v1 with AgentCraft v0 is 🤌 Managed up to 9 Claude Code agents with the RTS interface so far. There's a lot to explore, but it feels right. v1 coming soon

@skalskip92 global.espn.com/nba/story/_/id…

yo, nba people reached out to me about this tutorial.

I've been tracking this project for months though its evolution. Deserves praise and attention!

@JoePompliano ... and domestic.

The NFL’s first-ever regular-season game in Madrid will be played tomorrow at the newly renovated Santiago Bernabeu. The field is stored in an underground greenhouse with automated mowers, irrigation, and LED lighting. It's easily the NFL’s best international venue ever.

We believe this is the first documented case of a large-scale AI cyberattack executed without substantial human intervention. It has significant implications for cybersecurity in the age of AI agents. Read more: anthropic.com/news/disruptin…

@WagerWireCXO sits down with Kero Sports’ Tomash Devenishek to talk about the beauty of failing. Magical🪄✨ linkedin.com/posts/the-star…

Purchase was smooth (online after showroom visit + test drive). Incredible experience actually. X Plaid. Pickup was smooth but too impersonal IMHO. Walk in, get a key, told to go find it in the parking lot. Subsequent service at same location was very personable. Minus the outcome of being told that there's nothing they can do about acceleration shudders between 40-50mph and that I have to wait for a firmware update (been a year+ and nothing). My reco format to friends is always that I cannot see myself driving another car. I like other cars, but only aesthetically. Deep down one knows they won't compare. Converted a neighbor into the X from a Lexus.

Please reply to this post with any difficulties you may have had in trying to buy a Tesla. Our goal is for the purchase and delivery experience to be fast and simple, with accurate answers to your questions. The key test is that you would recommend it to a friend.

My first reaction to learning the details of the Luis Ortiz sports betting scandal/investigation was, "this is microbetting's fault." But after chatting with @tomashd about it, my feelings are more mixed.

@arlanr @nozomioai ✋ P.S. congrats and good luck on your journey from a fellow KZ-born founder.

holy shit. @nozomioai just 10x'ed my entire cursor workflow internally. agentic repo discovery, context enrichment, and extra memory for any codebase or documentation on the web. putting the final touches on it now. reply to this and I’d love to share it for free.