Bert Maher

@headinthebox It seems like in theory reorgs could solve problems (centralizing decision making in one person reducing communication and consensus costs) but in practice it doesn’t seem like they ever do

Textbook old school MSFT thinking: one more reorg will solve all your problems.

I am a proud American, and I’m proud to work at Anthropic. Labeling Claude a supply chain risk will only harm America’s lead in AI. It’s not un-American to support AI guardrails that protect our civil liberties.

I am deeply proud to work for this company.

@difficultyang Claude and I have so much in common 😂

claude gets so excited when analyzing pytorch profiler traces

@ViepliveeLee @giffmana 2.71828: Geminie

@tensorbert @giffmana I prefer Gemii 2.571415926535897932384626433832

Imagine they release Gemini 2.75 instead lol

@fermatslibrary One of the most common flaws of math textbooks is that they present only the logic, without the intuition. They give you the later, cleaned up version of the idea, which hides the way it was discovered.

This is all true, but Soumith is also one of the most brilliant strategic thinkers in the world. Some of us just fail a lot, dust ourselves off, and keep hacking the next day ☺️

This got me thinking that both int and FP math is “emulated” via a pretty complex set of transistors. I wonder how many gates/transistors it takes to implement an int8 fma versus an fp8, e4m3 fma

@tqchenml Fair point! My first guess would be “this path has regressed” but it’s also true that expectations are high, the hw is fast, and 10us can actually be substantial (depending on the work). If it’s the latter that’s rough, triton.jit is decently fast (need c++ launch maybe)

Two things worth noting, GPUs are likely getting faster. It is also good to ask about the upper bound possible. recently we get DSL host API overhead(excluding driver launch) down to 0.4us and around 1-2us(including driver launch) via tvm-ffi github.com/apache/tvm-ffi/

I’ve heard this complaint from a couple people recently, and I’m surprised because we optimized the launch path like a year ago and got it down to ~10us. There’s a now closed GitHub issue I filed with a microbenchmark - someone should run it, profile, and bring it down

@soumithchintala ❤️ It was great to have the chance to work with you, Soumith. I can’t wait to see what you do next

Leaving Meta and PyTorch I'm stepping down from PyTorch and leaving Meta on November 17th. tl;dr: Didn't want to be doing PyTorch forever, seemed like the perfect time to transition right after I got back from a long leave and the project built itself around me. Eleven years at Meta. Nearly all my professional life. Making many friends for life. Almost eight years leading PyTorch, taking it from nothing to 90%+ adoption in AI. Walking away from this was one of the hardest things I've ever done. But I'm leaving with a full heart. PyTorch handles exascale training now. It powers foundation models that are redefining intelligence. It's in production at virtually every major AI company. It's taught in classrooms from MIT to rural India. The tools I dreamed about making accessible? They are. The barrier to entry I wanted to lower? It's almost gone. To be clear, there’s so much more to do. As long as AI evolves at a breakneck pace, PyTorch will continue to play catch up. Obsessing over the yet-to-come sometimes makes us forget how much we’ve already done. To everyone who built this with me—who believed research should be joyful, that tools should be elegant, that open source changes everything—thank you. This wasn't my journey. It was ours. What's next for me? Something small. Something new. Something I don't fully understand yet. Something uncomfortable. I could have moved to something else inside Meta. But I needed to know what's out there. I needed to do something small again. I couldn't live with the counterfactual regret of never trying something outside Meta. It's very hard to leave. I probably have one of the AI industry’s most leveraged seats, I lead the software layer that powers the entire AI industry. Every major AI company and hardware vendor are on a speed dial. This kind of power is really hard to give up. But curiosity ultimately won out in my head. Keep making AI delicious and accessible. I'll be watching. Probably filing issues. Definitely staying involved. Is PyTorch going to be okay? I don't want to be doing PyTorch forever. I don't want to be like Guido or Linus— bound to a single thing for decades. Last November, coinciding with the birth of my daughter, I started planning my exit with Aparna. My goal was to leave PyTorch in a good and stable place. By this August, during the second half of my parental leave, I knew: Edward, Suo, Alban, Greg, John, Joe and Jana were ready. The team faced hard people, product, technical and organizational problems and didn’t feel the need to lean back on me to solve these for them (unlike in the past). The product story they crafted for the PyTorch Conference was coherent—really coherent. The things I'd flagged red were turning healthy. The project didn't need me anymore. Unlike 2020-2022 (when I stepped down to go do robotics and came back when Lin, Dima and Dwarak left), I have strong confidence that this time PyTorch is truly resilient. The most aligned culture carriers of PyTorch – Greg, Alban, Ed, Jason and Joe are at the decision table now, and people with strong value alignment – Suo, John and Jana have joined them at the table. And there’s a long list of equally value-aligned people willing to sit at the table should any of these people leave. There are many little things that make up my confidence on the people – John worked on Julia and open-source for a very long time (in fact we hacked a Torch.jl in 2015), Suo has been the strongest systems builder and strategic partner I’ve had for the past two years, and Jana worked on resilient core systems for a very long time, I’ve had long technical and organizational discussions with her over the past few months that give me confidence. And the product lineup and execution in 2025 should be sufficient evidence for any remaining doubt. I’m confident that this band of PyTorchers are going to do exceptionally well. PyTorch might change in flavor because I no longer impose my own taste from the top, but I’m confident that the values are going to stay intact and the product is going to be awesome. My time at Meta The early years of FAIR were absolutely magical. I was part of a small family of absolutely brilliant people building state-of-the-art AI out in the open. From working on GANs with Emily Denton, Rob Fergus, Leon Bottou, Martin Arjovsky and the (now legendary) Alec Radford to building Starcraft bots with Gabriel Synnaeve, to building the first FAIR Cluster with Howard Mansell, to working on object detection with Adam Lerer and Piotr Dollar, to building PyTorch. It was more fun than I can describe in words. 2015 and 2016 were probably the most productive and professionally enjoyable years of my life. I’ll probably romanticize this period of my life forever. When I joined FAIR, I had massive impostor syndrome, and the first 3 months were very very difficult. I can’t credit Andrew Tulloch enough for being the most thoughtful, kind and welcoming mentor, without whom I wouldn’t have made it. I’m so damn bullish for Meta just from the fact that he’s back. --- My time on PyTorch was special. I loved every part of building it—designing it, managing it, being the PM, TL, comms lead, doc engineer, release engineer, squashing bugs, growth hacking, turning it into a coherent product with hundreds of people, transitioning it to industry stakeholdership – the whole nine yards. To the core PyTorch team at Meta: the engineers, researchers, open-source maintainers, docs writers, CI infrastructure folks, hardware partners, the community builders. To the hundreds more inside and outside Meta—thank you. You turned a library into a movement. There are too many people to credit and thank, but I can't not mention Adam Paszke, Sam Gross, Greg Chanan, Joe Spisak, Alban Desmaison, Edward Yang, Richard Zou, Tongzhou Wang, Francisco Massa, Luca Antiga, Andreas Köpf, Zach DeVito, Zeming Lin, Adam Lerer, Howard Mansell and Natalia Gimelshein. And Schrep. They made the launch happen. And so many more people became centrally important later: Lu Fang, Xiaodong Wang, Junjie Bai, Nikita Shulga, Horace He, Mark Saroufim, Jason Ansel, Dmytro Dzhulgakov, Yangqing Jia, Geeta Chauhan, Will Constable, Briah Hirsh, Jane Xu, Mario Lezcano, Piotr Balecki, Yinghai Lu, Less Wright, Andrew Tulloch, Bruce Lin, Woo Kim, Helen Suk, Chris Gottbrath, Peng Wu, Joe Isaacson, Eli Uriegas, Tristan Rice, Yanan Cao, Elias Ellison, Animesh Jain, Peter Noordhuis, Tianyu Liu, Yifu Wang, Lin Qiao and hundreds more. It’s criminal of me to not take the space to list out everyone else I should be mentioning here. PyTorch is nothing without its people ❤️. The most joyful moments of building PyTorch was meeting users eager to share their happiness, love and feedback. I remember a grad student coming to me at Neurips 2017, in a slurring emotional voice he said he’d been trying to make progress on his research for 3 years but within 3 months of using PyTorch he made so much progress that he was ready to graduate. That moment made it tangible that what we do matters, a lot, to a lot of people, even if you don't constantly hear from them. I do miss the intimacy of the PyTorch community, with a 300 person conference that felt like an extended family gathering, but I feel that’s a small price to pay considering the scale of impact PyTorch is truly having today – yes the Conference is now 3,000 people where market-moving deals get brokered, but it’s helping orders of magnitude more people to do their best AI work. I miss the intimacy, but I'm proud of that growth. --- To Mark Zuckerberg and Mike Schroepfer, who believed that open-sourcing is fundamentally important and is a sound business strategy. This is so hard to understand for most people within the course of business, but we’ve run lock-step on this strategy without ever having to discuss it. Without you two, neither FAIR nor PyTorch would’ve happened. And those mean so much to me. To Yann LeCun and Rob Fergus, for building the magical early FAIR that I so revere. To Aparna Ramani, a leader that I find so rare at Meta in her ability to hold a really high bar for the org, technically brilliant with the span to discuss deep infra systems and industry-strategy within the same conversation and for being an absolute execution-machine! I’ve learned so much from you. To Santosh, Kaushik, Delia, Oldham and Ben for being so welcoming to Infra. For someone coming over from FAIR with a wildly different culture, you all made me feel at home and made me part of the family, and thank you for that. To all my managers who've championed me through the PSC video game – Serkan, Howard, Jerome, Abhijit, Yoram, Joelle, Aparna and Damien – I owe you a lifetime of drinks. --- Signing off for now. —Soumith

@headinthebox Might it be more tractable to verify that two implementations match, than to come up with an optimized implementation of a simpler one?

Perfect example of why I got disappointed with program verification. Too often the post-condition is just a pure functional version of the imperative implementation. Why bother even writing the imperative code, assuming a decent compiler will optimize the former into the latter under the covers for you.

It would be kind of cool if torch.compile could be used as a context manager, like: ``` some_custom_kernels() with torch.compile(): # do a bunch of easy pointwise stuff more_custom_kernels() ```

@ScottWolchok @marksaroufim Sometimes I think the the rows-vs-columns framing is kind of unhelpful. I sometimes think about matmul with rhs transposed, so you have an [m,k] matrix and and [n,k] matrix, and you end up with an [m,n] of all the dot products over k. (Which is kind of what nn.Linear does)

@marksaroufim @tensorbert "there is a cube filled with products, yes we sum those along exactly one dimension of the cube, no we are not just summing all the things" is a lot more helpful than the high-school "OK you go across the rows of the one and down the columns of the other, who knows why"

Read this to the end — the last section is mind-blowing

I am really enjoying using Claude Code with Sonnet 4.5! It's super smart, and super fast!

Introducing Claude Sonnet 4.5—the best coding model in the world. It's the strongest model for building complex agents. It's the best model at using computers. And it shows substantial gains on tests of reasoning and math.

😍

@matt_dz @davorVDR Oh man can’t believe I forgot Helion in the list. And it compiles to triton (or at least did last I looked) so it’s turtles all the way down

@davorVDR @tensorbert and more coming, e.g., github.com/pytorch/helion

lol, there is quite the explosion of kernel DSLs lately (triton, tilelang, gluon, TLX, cuteDSL, cuTile, …) And honestly as much as I love TLX and want it to succeed, I think the next big kernel programming language might be… natural, human language

@vinodg Indeed - but I think that language can increasingly become more intuitive and flexible

@tensorbert You’ll still need to tell it what you want.