Learn 𝗚𝗣𝗨 𝗽𝗿𝗼𝗴𝗿𝗮𝗺𝗺𝗶𝗻𝗴 from the ground up: begin with Numba for low-level control, then progress to Triton to write high-performance kernels in a Python-like language. A hands-on Jupyter notebook to get you started quickly.

Performance Hints Over the years, my colleague Sanjay Ghemawat and I have done a fair bit of diving into performance tuning of various pieces of code. We wrote an internal Performance Hints document a couple of years ago as a way of identifying some general principles and we've

some recent reads from this month that I've learned from and that are pretty cool 1. Inside NVIDIA GPUs: Anatomy of high performance matmul kernels aleksagordic.com/blog/matmul 2. Triton Flash Attention Kernel Walkthrough: The Forward Pass nathanchen.me/public/Triton-… 3. This guy

This work by Cursor is, I think, the coolest AI breakthrough since GPT-4. (And there are plenty of candidates!) simonwillison.net/2026/Jan/19/sc…

🚀 MLSys 2026 Contest - NVIDIA Track is LIVE! Registration is now open for the FlashInfer-Bench Challenge! Submit high-performance GPU kernels for cutting-edge LLM architectures on NVIDIA Blackwell GPUs. Three Tracks * MoE (Mixture of Experts) * DSA (Deepseek Sparse Attention)

x.com/i/article/2014…

Nice work, Abhishek Maiti! 🙌 This kind of write-up helps more folks understand the internals and start building. If you try something new after reading, consider upstreaming it to vLLM. We’d love to collaborate.

This paper on the current state of AI Agents is worth reading. Main points: 1. Add memory to agents. 2. Build agents as loops, not pipelines. 3. Go for RL only after the Agent's behavior is reliable. 4. Specify which tool to use when (don't dump 50 tools into a prompt and

not having a GPU is not an excuse to learn CUDA/triton. > sign in to platforms like leetgpu/tensara. > make an account > solve challenges, look at other solutions , read docs, get better.

gold mine of an account - highly recommend checking out his content!

x.com/i/article/2028…

Bro just reverse engineered how Codex actually works. Worth read !!

Maintaining separate attention kernels for every GPU platform doesn't scale. The vLLM Triton attention backend takes a different approach: ~800 lines of Triton, same source code across NVIDIA, AMD, and Intel GPUs. On H100, it matches state-of-the-art attention performance. On

- 230 training runs - 1,623 GPU hours (67 B200 days) - 76 TB of training data - a 2x faster model Every paper said it can't be done. Quantization Aware Distillation made it possible.

My timeline feels so blessed today. Full of deep learning maniacs writing cool articles like this. 👀

x.com/i/article/2031…

If you guys aren't following the CEO of Cerebras-you should be. One of the brightest guys in the industry. Great writeup for the technical analysis guys who know nothing about AI. I'm somewhere in the middle and learned a lot here. Andrew Feldman would love to hear your

Very hyped by the new Cohere Transcribe model 🌍 Works surprisingly well on bad quality audio when the mic doesn't cooperate. 2B params, 14 supported languages and it's Apache 2.0. try the official Hugging Face demo ⬇️

NVIDIA & Unsloth dropped one of the best practical guides on building RL environments from scratch, filling gaps most tutorials skip. Covers: - Why RL environments matter + how to build them - When RL beats SFT - GRPO & RL best practices - How verifiable rewards & RLVR work