Naol Negassa

I heard this somewhere, and it’s really stuck with me. -In times of chaos, the people who are willing to provide innovation, stability and opportunity to the people around them, win. Be one of those people.

This entire cyberpunk world was built by a single creator. 100 million Gaussian splats, with nearly every surface and structure generated in Marble. We’re entering an era where individuals can build entire worlds.

Introducing Alterego: the world’s first near-telepathic wearable that enables silent communication at the speed of thought. Alterego makes AI an extension of the human mind. We’ve made several breakthroughs since our work started at MIT. We’re announcing those today.

Today, we're announcing @episteme, a new type of R&D company that recruits exceptional scientists to pursue high-impact ideas. Science isn’t bottlenecked by the availability of talent, but by places where they can do their best work. Scientific progress has driven human flourishing: extending lifespans, lifting billions from poverty, and expanding our understanding of the universe. But history is littered with transformational ideas that were overlooked in their time. That problem is still acute today: too much promising talent remains uncultivated, and remarkable ideas die in the lab or are filtered out by misaligned incentives. Today, scientists face suboptimal paths for translating their research into impact: academia is famously risk-averse and incentivizes publications and winning grants vs. translational research. Industry is too often focused on short‑term incentives. And startups lack the substantial capital, expertise, and complex infrastructure needed to deliver long-term scientific progress. On top of that, recent funding cuts in the US mean the overall supply of ideas is decreasing. Put together, the global scientific production system is operating at a fraction of its capacity. How Episteme operates is different: we identify great scientists who can meaningfully benefit humanity, but who aren’t supported efficiently within traditional institutions today. Researcher by researcher, we work with them to determine the bespoke resources, operational support, and environmental conditions to execute on their research. We bring them together in-house, and provide those resources to ensure that their breakthroughs are deployed for real-world impact. We’ve already assembled an amazing team of operators, ranging from the Gates Foundation, DeepMind, ARPAs, DoE – just to name a few – and researchers who are pursuing important problems across physics, biology, computing, and energy. Our team has spoken to hundreds of researchers across disciplines and geographies to understand the limitations they’re facing and what can be done better, and designed Episteme for them. We’re backed by individuals like @sama, Masayoshi Son, and other long-term partners who share our mission of enabling ambitious science for tangible human impact. About me: I started working as a researcher 9 years ago, on problems ranging from AI-driven drug discovery to developing brain-machine interfaces. It was that experience that led me to realize that so many scientists with great potential to change the world don’t have access to opportunities equal to their capacities. @sama and I believe that much better science should happen for humanity, and that a new engine is needed to support that. We decided to cofound Episteme together, and I am incredibly grateful for Sam’s unwavering support as a thought partner and founding investor. Our conviction is that by supporting the right people with the right incentives, we're set to generate breakthrough discoveries to benefit humanity. We cannot rely on the course of history to shape scientific progress; we need to proactively shape the system by supporting the most talented people with the right resources and incentives.

In the first year of my PhD, I formalized the independence of the continuum hypothesis in Lean. It was considered a breakthrough at the time, taking ~12 months and 20K lines of code. With Gauss, we finished Strong PNT with 25K LOC in 3 weeks. There are two really remarkable things about this result. The first is that there is no single human who is really familiar with this artifact. The second is that there is something truly futurebound about working with something that can go off for a thousand minutes at a time rather than a hundred, to do work that maybe only a few dozen humans on earth could have done given several days. While @morph_labs will soon enable anyone to scale agents with computer workspaces to that level, I am incredibly excited by the work to be done at Math, Inc. - humanity advances when we invent trustworthy interfaces; when we invented compilers and higher-order programming languages, we discovered an entire universe of software. Autoformalization will do the same for mathematics, and with @ChrSzegedy and @jdlichtman we will find verified superintelligence on the other side.

The #EEGManyLabs website is now live: eegmanylabs.org A home for our global effort to test the replicability of influential EEG findings, share resources, improve methods in cognitive neuroscience, and grow an open, connected community.

Today, we're announcing our $50M Series A and sharing a preview of Ember - a universal neural programming platform that gives direct, programmable access to any AI model's internal thoughts.

Today, we are excited to announce Thinking Machines Lab (thinkingmachines.ai), an artificial intelligence research and product company. We are scientists, engineers, and builders behind some of the most widely used AI products and libraries, including ChatGPT, Character.ai, PyTorch, and Mistral. Our mission is to make artificial intelligence work for you by building a future where everyone has access to the knowledge and tools to make AI serve their unique needs. We are committed to open science through publications and code releases, while focusing on human-AI collaboration that serves diverse domains. Our approach embraces co-design of research and products to enable learning from real-world deployment and rapid iteration. This work requires three core foundations: state-of-the-art model intelligence, high-quality infrastructure, and advanced multimodal capabilities. We are committed to building models at the frontier of capabilities to deliver on this promise. If you’re interested in joining our team, consider applying here: 6wajk07p.paperform.co

A brilliant idea isn’t a fact—until it is. Many groundbreaking discoveries seem obvious only in hindsight, once they unify a web of seemingly isolated facts into a general principle. Before we connected the dots between evolution, genetics & material science, silk was just a thread, proteins were just biological molecules, and genes were just codes. But once we saw their relationships, we unlocked deep truths about how nature builds materials at every scale. What If AI Could Think in Relationships Instead of Just Memorizing? Most AI today doesn’t work this way. It merely predicts the next token, unaware of whether its own output is meaningful, correct, or groundbreaking. They: ❌ Lack true reasoning—they do not verify if their responses make sense. ❌ Canot correct themselves—once they generate something, they have no mechanism to reflect and refine their own ideas. ❌ Do not connect ideas deeply—they retrieve, not discover. 💡 SciAgents does something different. Rather than treating knowledge as isolated facts, it builds a massive relational graph, connecting every concept and idea to others. Then, a team of AI agents—much like a group of researchers—explores this graph, not just by taking the shortest path between ideas, but by wandering through unexpected links. How SciAgents Reasons over Graphs ▶️Instead of taking the shortest path between two ideas (which can be too direct & limiting), SciAgents samples diverse paths through a powerful algorithm that explores ever-growing sets of diverse waypoints. This allows it to natively explore broader, richer relationships—leading to unexpected discoveries. ▶️For example, to explore the connection between silk and energy efficiency, SciAgents didn’t just look at direct links. It uncovered intermediate concepts like biocompatibility, multifunctionality, and structural coloration, revealing new ways to design bioinspired materials that human researchers might have overlooked. Why does this matter for building better AI for science and beyond? 1⃣Generalization is the key to intelligence. Memorization alone won’t get AI to true reasoning—but structuring knowledge in a relational way can. 2⃣SciAgents goes beyond predicting words. It constructs ideas by mapping biological blueprints—from genes encoding proteins to evolutionarily refined materials like silk—and extrapolates new designs for synthetic biology and materials science. 3⃣It refines its own outputs. Rather than passively generating text, SciAgents’ multi-agent system debates, critiques, and improves hypotheses, making its discoveries deeper and more reliable. 🌍 Graph-based reasoning plus multi-agent collaboration is not just a better way for AI to think—it’s likely on a critical path towards AGI. The ability to form deep, structured insights from sparse information is what separates mere computation from true intelligence. A. Ghafarollahi, M.J. Buehler, SciAgents: Automating Scientific Discovery Through Bioinspired Multi-Agent Intelligent Graph Reasoning, Adv. Materials, DOI: 10.1002/adma.202413523, 2025

I built a realtime multimedia tutor that: 🎨 Generates visuals in realtime (diagrams, photos, sketches) while it speaks 👁️ Sees what you see (via camera/screenshare) to help in context ✍️ Writes key facts, equations, details on the screen — no rambling monologues Watch me point my camera at my ailing houseplant 🌱, and it simultaneously writes out the biology facts, illustrates the issue on the leaves, and talks me through how to fix it. I can't open it for everyone just yet, but I want to v soon. In the meantime, reply with demos you want to see, or DM me if you want to try the preview!

Thrilled to join @A_ACollective, a global community for emerging investment professionals driving growth in the African tech and investment ecosystem. Looking forward to collaborating, learning, and contributing to this impactful network! 🌍

The secret to true AGI will not be a single omnipotent model... It'll be a series of well-defined ontologies for specific problem spaces with many AI models + code to strike the right balance of intuition + logical guardrails for agentic reasoning, planning and decision-making

Piramidal (W24) is building a foundation model for the brain. Imagine ChatGPT but for the “language” spoken by your neurons. Their first product is a copilot for epilepsy diagnosis that helps neurologists get insights around a patient’s EEG, in real-time: ycombinator.com/launches/KVc-p…

I’m starting to believe that “write on the Internet, even if no one reads it” is underrated life advice. it doesn’t benefit other people necessarily, it benefits *you* bc the people who do find/like your writing and then reach out are so much more likely to be compatible with you

With @Harvard, we built a ‘virtual rodent’ powered by AI to help us better understand how the brain controls movement. 🧠 With deep RL, it learned to operate a biomechanically accurate rat model - allowing us to compare real & virtual neural activity. → dpmd.ai/3RobU7e

More and more companies are going to get into "managed agents" soon where they run the compute and provide devs stateful APIs/orchestration. The buzzword of 2024/5 is going to be "Agent API." You're going to see announcements like "Introducing the Salesforce Agent API" and "the new Notion Agent API" , etc. The way these offerings will work: they'll use frontier models + sequential tool calls to offer task planning APIs to developers. Then, devs will be able to schedule async/step-function behavior and poll for updates within their own products. They'll likely be 2 flavors of this. They'll be monolithic services where the company runs the compute and the scheduler themselves for devs. Then, they'll be headless offerings, where planning, scheduling, and execution are decoupled.

@joaomdmoura Nice! Did you use them to update the docs to the new version?

Now I have a crew that can re-write all crewAI docs making it better and adding new features to it 😎👉👉 maybe I should record a demo on this one as well?

@ashpreetbedi If I’m not mistaken, you sort of already handle this with @phidatahq , right? Or maybe I’m missing something here?

Excited to try out FireFunction!

@intrstllrninja @NousResearch Got it, thanks!

@NaolNegassa @NousResearch you don't need to convert to openai function if it's just json generation -- model can generate json object out of the box provided with pydantic schema

JSON mode with Mistral-7B has a pass rate of 80% 🔥 Mistral-7B base was finetuned on a mix of mini Hermes, function calling, json-mode and agentic datasets. stay tuned for struct models & datasets from @NousResearch 🥽

@intrstllrninja @NousResearch Very cool! Would you still need to do stuff like convert_to_openai_function for tools in langchain?

@NousResearch besides function calling and json mode, we have also trained this model on agentic reasoning frameworks with examples from LLM frameworks like LangChain and Instructor so the model is capable of in-context reasoning and output a json object with reasoning and final response