AI300
  1. Aidan Smith / Cofounder of Flapping Airplanes
    1021.
    incredibly proud of our work on the kitten virtual machine.›
    LLM score 92 · 3 months ago
  2. Cyris Kissane / Researcher at Flapping Airplanes
    1022.
    come build agi infrastructure with us @flappyairplanes ! https://t.co/827cRlPbSf
    LLM score 75 · 3 months ago
  3. Benjamin Spector / Cofounder of Flapping Airplanes
    1023.
    Fun to share a little taste of some of the things we've been working on with @sequoia https://t.co/C09n0TmrAk https://t.co/XbJasJHSr2
    LLM score 75 · 3 months ago
  4. Michael Elabd / DeepMind Researcher
    1024.
    AI in legal feels like it’s reaching an important inflection point!›
    LLM score 82 · 3 months ago
  5. Ronak Malde / DeepMind Researcher
    1025.
    Long horizon tasks in many knowledge work domains are still an entirely unsolved problem.›
    LLM score 92 · 3 months ago
  6. Cameron Wolfe / Researcher at Netflix
    1026.
    Here's another great paper on creating realistic / modern evals for agents with a slightly different flavor: https://t.co/nMSHyu8NZY›
    LLM score 92 · 3 months ago
  7. Ben Burtenshaw / Hugging Face Researcher
    1027.
    this blog post is a thorough deep dive into the rl environment ecosystem.›
    LLM score 85 · 3 months ago
  8. Leandro von Werra / Hugging Face Head of Research
    1028.
    You can just do things (instead of onboarding)! At Hugging Face we skip onboarding and let new hires just build and release projects.›
    LLM score 92 · 3 months ago
  9. Christian Szegedy / ex xAI Cofounder
    1029.
    Some things never change: Gary just doesn't fit in distribution.›
    LLM score 92 · 3 months ago
  10. Sander Dieleman / DeepMind Research Scientist
    1030.
    My first blog post in over a year is a deep dive on flow maps🗺️, or how to learn the integral of a diffusion model to enable faster sampling and several other cool tricks.›
    LLM score 95 · 3 months ago
  11. Ronak Malde / DeepMind Researcher
    1031.
    One of the biggest concerns for AI researchers today should be that we are benchmaxxing without a purpose.›
    LLM score 92 · 3 months ago
  12. Sherwin Wu / OpenAI API, Head of Engineering
    1032.
    ChatGPT for Excel and Google Sheets is now out of beta! ›
    LLM score 75 · 3 months ago
  13. Cameron Wolfe / Researcher at Netflix
    1033.
    The way we evaluate LLMs has to change as we move towards better and more autonomous agents.›
    LLM score 92 · 3 months ago
  14. Yoshua Bengio
    1034.
    Thank you @UN Secretary General @antonioguterres for an excellent discussion on the current state of AI capabilities, and the urgent need for robust global governance.›
    LLM score 85 · 3 months ago
  15. Merve Noyan / Hugging Face ML Engineer
    1035.
    Gemma 4 just got a massive speed-up with MTP drafters ⚡️ ›
    LLM score 92 · 3 months ago
  16. Igor Babuschkin / Cofounder of xAI
    1036.
    sglang is the best inference framework out there.›
    LLM score 75 · 3 months ago
  17. Ethan Shen / Ai2 Researcher
    1037.
    Imo e2e repository creation is the last barrier towards fully autonomous coding agents.›
    LLM score 92 · 3 months ago
  18. Ben Burtenshaw / Hugging Face Researcher
    1038.
    Introducing the context course: a free course on doing ML with agent context.›
    LLM score 85 · 3 months ago
  19. Merve Noyan / Hugging Face ML Engineer
    1039.
    I forked openclaw to build a small local rescue agent that debugs whenever the agent is down›
    LLM score 92 · 3 months ago
  20. Sherwin Wu / OpenAI API, Head of Engineering
    1040.
    Even though GPT-5.5 is 2x more expensive than GPT-5.4, you only pay ~49-92% more, not 100% more – because it uses fewer tokens to get the job done.›
    LLM score 92 · 3 months ago
  21. Aidan McLaughlin / OpenAI Research Scientist
    1041.
    the idea of a singleton or a coherent agi persona that survives into the late 20s seems absurd to me.›
    LLM score 92 · 3 months ago
  22. Kevin Murphy / DeepMind Research Scientist
    1042.
    Major update to my "Bayesian Linguistic Forecasting" paper! I have now tried it on 5 different LLMs: Gemini 3.1 Pro, Gemini 3 Flash, Sonnet 4.6, GPT 5.4 and Kimi K2.5.›
    LLM score 92 · 3 months ago
  23. Keller Jordan / OpenAI Researcher
    1043.
    Three new modded-NanoGPT optimization benchmark results, all of them using NorMuon, have near-concurrently improved the benchmark record from 3325 to 3250 steps.›
    LLM score 92 · 3 months ago
  24. Jack Clark / Anthropic Cofounder
    1044.
    I've spent the past few weeks reading 100s of public data sources about AI development.›
    LLM score 92 · 3 months ago
  25. Ben Burtenshaw / Hugging Face Researcher
    1045.
    Hey Ben, where are all the courses? In 2026, I’ve cut down a lot on new courses for Hugging Face.›
    LLM score 85 · 3 months ago
  26. Andrew White / Edison Scientific Cofounder
    1046.
    How can you tell if you're on the latest chatgpt image generator? It still cannot draw molecules for me (but everything else is quite good).›
    LLM score 92 · 3 months ago
  27. Christian Szegedy / ex xAI Cofounder
    1047.
    That's how early we are in the AI economy — the 1995 of IT.›
    LLM score 92 · 3 months ago
  28. Keller Jordan / OpenAI Researcher
    1048.
    New modded-NanoGPT optimization result: GH:WilsonCWu has matched the current record using an adaptive Muon variant.›
    LLM score 92 · 3 months ago
  29. Keller Jordan / OpenAI Researcher
    1049.
    Notable modded-NanoGPT optimization result: GH:WilsonCWu has matched the current record using an adaptive Muon variant.›
    LLM score 92 · 3 months ago
  30. Cameron Wolfe / Researcher at Netflix
    1050.
    Gemma-4 has received more attention than prior generations of Gemma, but I still somehow feel like these models are so underrated.›
    LLM score 82 · 3 months ago
More