I had a lot of fun writing a new blog: Optimizing NSA for TPUs There's an accompanying colab notebook with it too! I hope this helps people tinker with NSA in JAX + TPU kernels with Pallas

📣Call for Blog Posts at #ICLR2026 ICLR 2026 Following the success of the past iterations, we are opening the Call for Blog Posts 2026! iclr-blogposts.github.io/2026/about/#a-… Please retweet!

Hey Y Combinator Garry Tan someone is impersonating you guys, spamming and scamming GitHub accounts github.com/y-combinatoror…

Yi Ma Academia is far too infiltrated with industrial incentives. The upside is there's more funding and bigger class sizes. The downside is that industry incentivizes "showmanship" over true "scholarship".

People who can abstract, develop, design and serve systems end to end are increasingly rare. Knowing how to do that turns you from a "generalist" to the MVP with 100x impact.

If you're interested in LLMOps and GPU programming and you're not subscribed to ModelCraft, you should be. email.mg-d1.substack.com/c/eJxskDtuwzAQ…

I've got a few free monthly subscriptions for Abi Aryan ModelCraft Substack Message me if you're interested.

Do you remember when you joined X? I do! #MyXAnniversary

Sort of rearranging some chapters on the GPU Engineering book.

People after apple dev con - apple is the only hardware company that cares about us. People after open ai dev day - openai is winning, apple is so dead 😂 People after gtc - nvidia is the only one winning, everyone else is so dead 😆

I created an `awesome-list` on Github with some of the beginner resources on GPU Engineering. Link: gpuengineering.com Feel free to contibute and add more - just getting started on collating resources.