Joe Isaacson
189 posts
Joe Isaacson
@Jisaacso
torching pys @pytorch
Seattle, WA Katılım Temmuz 2012
848 Takip Edilen467 Takipçiler
After 5 amazing years, I’m leaving the PyTorch team at Meta. I did my best work there and got to work with some of the smartest, most OSS pilled engineers in the industry. More soon on what’s next: still systems, still OSS (but not everything), a smaller team with a lot of GPUs
English
@elliotarledge @PyTorch The focus of our initial release was correctness to the precision level of KernelBench. We'll continue hammering on both accuracy and performance next!
English
KernelFalcon achieves 100% correctness across all 250 KernelBench L1–L3 tasks through a deep agent architecture that structures the problem instead of prompting harder.
The system combines hierarchical task decomposition, deterministic orchestration, grounded execution, and parallel verification to generate GPU kernels that compile to PTX, execute on real hardware, and preserve PyTorch semantics.
💡Read our latest blog from @LaurawlyLaura and collaborators at Team PyTorch:
hubs.la/Q03RYkZq0
#PyTorch #KernelFalcon #AIInfrastructure #OpenSourceAI
English
@giffmana Shout out to @ralfgommers and @quansightai ‘s partnership with @PyTorch team last year to ship compiled numpy!
English
Oh wow, did you guys know that torch.compile can compile numpy code? And even run it on GPU?
This is pretty neat for all kinds of "surrounding" code besides the model (like evals and fancy metrics) that I used to do with numba/numexpr (cuz CPU-XLA was pretty meh).
Poll below
English
Joe Isaacson retweetledi
Would you rather have better documentation for PyTorch prototype features but more BC breakages or secret APIs until we have more confidence?
English
🚀🚀🚀 come work with PyTorch on GPU kernel optimization, distributed training/inference/post-training, compiler design, ABI stability
Mark Saroufim@marksaroufim
If you’re excited about optimizing code that runs equally well on a single or thousands of GPUs and if you have the ability to submit a single substantial PR to a major OSS library, we want you on the PyTorch team - especially if you’re early in your career.
English
Excited to share the GPU MODE kernel authoring competition! We’re stoked to see the creative solutions people come up with to push bounds of performance. And, all for a good cause — towards a public dataset of performant kernels 😁
GPU MODE@GPU_MODE
Write a fast kernel and run it on Discord. See how you compare against the best! If you're familiar with Leetcode, Kaggle or Codeforces then this should feel right at home
English
@AsimovBio cooking over the holidays. Just launched their second essays on genetic engineering, written in physical DNA and sealed in a stainless steel capsule. Lasts 10k years without degradation 🤯
Alec Nielsen@alectricity
Eat your heart out, Gutenberg @AsimovPress book #2, encoded in DNA, available now for pre-order. One of the coolest projects I've ever seen unfold. Huge thanks to @catalogtechinc, @imagene_biotech, and @plasmidsaurus for making this possible, and @jasonjoyride for the epic vid
English
Great talk on the importance of RLEF and CoT for coding generation performance
Gabriel Synnaeve@syhw
Just gave a talk on "Grounding LLMs in Code Execution" at the NeurIPS Hacker-Cup AI Competition, here are the slides docs.google.com/presentation/d…
English
@anneouyang from @Stanford talking on KernelBench, a standardized benchmark for kernel generations. Level 1-4 problems challenging sota LLMs to generate performant GPU kernels.
English
@simran_s_arora from @HazyResearch on GPU kernel generation, project 🍿and the challenge of AI mapping algorithms to hardware architecture. At @NeurIPSConf Hacker AI Cup
English
@neal_wu from @cognition_labs on the future of competitive programming as a litmus test for AI, at @NeurIPSConf Hacker Cup session
English
David Harmeyer on Facebook’s history of Hacker Cup. How to add N flights to N airports to minimize travel distance.
English
Stoked for the Hacker AI Cup Workshop to start at @NeurIPSConf ! Speakers @weiwei_msr @syhw @neal_wu @simran_s_arora @MishaLaskin @thomascapellajp @marksaroufim ! Come join, room 210
English
Congrats on the launch of 🌽 KernelBench! We're excited to partner with 🍿 Popcorn to lower the barrier to GPU programming!
Simon Guo@simonguozirui
We are not the only ones thinking about this direction. The vision was nicely outlined by @karpathy in his talk (see picture below, credit to karpathy and @accel), and we found many others also eager to attempt this. Project 🍿: gpu-mode.github.io/popcorn/ in the GPU Mode server is a community effort towards this goal of LLMs generating performant GPU kernels. We hope KernelBench could be a valuable contribution for the community, to help understand current capabilities and serve as a playground to evaluate various techniques. If you are interested, we would love to hear your thoughts and welcome you to join the mission! Link to the relevant part of Karpathy’s talk: youtu.be/aR6CzM0x-g0?fe…
English
If you could change one thing about PyTorch what would it be?
English
