PyTorch, every mention
35 scenes · ← back to PyTorch
tap a year for its mentions
every year anyone Matt Turck 28Lin Qiao 12Jerome Pesenti 11Lukas Biewald 7Dylan Patel 5Clement Delangue 4Zico Kolter 3Douwe Kiela 2Dave Burgess 2William Falcon 1
Verbatim, from the transcripts: the passages where PyTorch comes up
Inside Nemotron & NVIDIA’s AI Lab | Bryan Catanzaro
- ▶ 14:30 Bryan Catanzaro Uh, first Copperhead, which was a, it was a, a Python, uh, embedded language that, um, compiled to the GPU, um, which I think foreshadowed a lot of things in TensorFlow and, and PyTorch.
OpenAI Board Member Zico Kolter: Modern AI Is Just 200 Lines of Code
- ▶ 1:11:40 Zico Kolter You use PyTorch, but you build one from scratch that, you know, can be a chatbot.
- ▶ 1:13:35 Zico Kolter It uses PyTorch, but it doesn't even use any of the prebuilt layers. 2 times in the scene
Voice AI’s Big Moment: Top Researcher on Why Everything Is Changing (Neil Zeghidour, Gradium AI)
- ▶ 14:39 Neil Zeghidour So for the anecdote, I did my coding interview with Sumit Chintala, who then invented PyTorch, and now I think he's the CTO of Thinking Machines.
Dylan Patel: NVIDIA's New Moat & Why China is "Semiconductor Pilled”
- ▶ 11:51 Dylan Patel Like most people write PyTorch. 4 times in the scene
- ▶ 38:49 Dylan Patel BLM and PyTorch, SGLang, and like all of these other like libraries and things that are just like, you know, and it goes to low level software, especially, right?
Why This Ex-Meta Leader is Rethinking AI Infrastructure | Lin Qiao, CEO, Fireworks AI
- ▶ 0:27 Matt Turck We talked about how her experiences with PyTorch and Meta led to the creation of Fireworks. 2 times in the scene
- ▶ 2:47 Matt Turck Maybe a few words on your story, your background, in particular, uh, you were one of the key people, uh, on PyTorch, uh, at, uh, Meta, or Facebook at the time? 3 times in the scene
- ▶ 7:17 Lin Qiao Um, at that time, uh, PyTorch externally, it's a open source project. 2 times in the scene
- ▶ 8:00 Lin Qiao It started from research labs, and the fun fact is OpenAI, uh, switched to use PyTorch fully. 5 times in the scene
- ▶ 14:55 Matt Turck And just to close on PyTorch, and then we'll go into, uh, fireworks in, uh, in, in detail, um, as, as a, as a passing thought, and that sort of echoes a conversation I was, uh, just having most recently with, um, Dao, um, from Contextual… 4 times in the scene
- ▶ 16:20 Matt Turck and then PyTorch influence the idea for fireworks in the first place? 8 times in the scene
- ▶ 22:30 Lin Qiao Um, and that's basically a customized PyTorch inference infrastructure we set up for JNS specifically. 2 times in the scene
- ▶ 44:36 Lin Qiao Right, so we have been invest in, um, overall the speed and cost optimization, there are two sides of the same coin, shall we say, uh, for a long time, um, and from the PyTorch days, and we, we actually have a lot of, uh, customized kernel. 2 times in the scene
- ▶ 56:18 Lin Qiao Um, I'm very bullish on open models because I have seen the power of open source that PyTorch has been able to leverage.
Top AI Researcher on GPT 4.5, DeepSeek and Agentic RAG | Douwe Kiela, CEO, Contextual AI
- ▶ 37:35 Douwe Kiela Lama and PyTorch and yeah, so much, right? 2 times in the scene
The $4.5B Platform Driving the Open Source AI Revolution | Clem Delangue, CEO, Hugging Face
- ▶ 29:54 Clement Delangue From, ah, from Google, but, ah, but it kind of sucks because it's in, ah, TensorFlow, and, and most people today wanna use PyTorch. 4 times in the scene
Hollywood Producers FREAK OUT over AI | Cris Valenzuela, CEO of Runway
- ▶ 3:38 Cris Valenzuela Um, Image was working really well, not really well, sorry, um, like GANs were like the state-of-the-art thing at the time, but for me it wasn't really, um, like one algorithm or one technique, it was more when I started, like, delving into…
Moonhub’s Nancy Xu Unveils the AI Recruiter That’s Beating LinkedIn
Reinventing Search with AI: Richard Socher on Building You.com & the Future of Google
- ▶ 2:04 Richard Socher This was, you know, before there was high torch, before it was TensorFlow, uh, there are a few sort of smaller frameworks.
Startup to Industry Standard: Lukas Biewald Explains How W&B Scaled MLOps for OpenAI, NVIDIA & More
- ▶ 7:19 Lukas Biewald You know, at the time, like PyTorch, I don't think PyTorch ever really had a great TensorBoard integration, so we wanted to integrate with more stuff and just make it friendlier.
- ▶ 47:22 Lukas Biewald Um, and of course we see more lightning now and, and I mean, obviously PyTorch is like, I mean, I think PyTorch is like phenomenal. 6 times in the scene
Build and Deploy AI with Pytorch | Lightning AI Founder & CEO, William Falcon
- ▶ 0:13 Matt Turck So you are the CEO of, uh, Lightning AI, the platform to train, deploy, and build AI with PyTorch, Lightning Fast. 3 times in the scene
- ▶ 1:07 Matt Turck So what is PyTorch, and what was your journey from PyTorch to, uh, Lightning? 6 times in the scene
- ▶ 25:33 William Falcon The other thing that I haven't mentioned is, we have an in-house PyTorch team, so we hired a lot of the core leads from Meta, so, um, the, you know, the person who wrote a lot of the compile stuff that you're seeing here, the linear…
A Conversation on The State of AI | Melanie Kambadur, Meta & Gideon Mann, Bloomberg
- ▶ 0:57 Gideon Mann In the form of things like, uh, Torch and PyTorch.
Fireside Chat: Dave Burgess (Head of Data Engineering, Pinterest) w/ Matt Turck (Partner, FirstMark)
- ▶ 16:07 Dave Burgess We use PyTorch, scikit-learn.
- ▶ 22:38 Dave Burgess And so we provide the, the glue and the underlying infrastructure, uh, with TensorFlow and PyTorch and all the, all the other components available to our internal customers.
Fireside Chat: Chip Huyen with Matt Turck (Partner, FirstMark)
- ▶ 7:01 Chip Huyen So I think by pretty, in, in, in, in, in, it's pretty much the competition is between like Pytosh and TensorFlow.
Fireside Chat: Savin Goyal (ML Infra team (Metaflow), Netflix) with Matt Turck (Partner, FirstMark)
- ▶ 16:18 Savin Goyal I want to see if they are on the TensorFlow versus PyTorch, like a whole bunch of different engineering related questions that,
Fireside Chat: Alok Gupta (Head of Data Science & ML, DoorDash) with Matt Turck (Partner, FirstMark)
- ▶ 5:40 Matt Turck So just, uh, again, reading through some of the stuff I've, I've seen XGBoost, LightGBM, CatBoost, obviously TensorFlow and PyTorch on the, on the neural network, um, side. 2 times in the scene
Fireside Chat: Jerome Pesenti (Head of AI, Facebook) with Matt Turck (Partner, FirstMark)
- ▶ 1:44 Jerome Pesenti Uh, we are the developers of PyTorch. 3 times in the scene
- ▶ 23:32 Jerome Pesenti So obviously PyTorch is, uh, one of our flagship project at the moment, and we really trying to build a community around this, but not only that, we're really trying to incentivize people to share their models. 4 times in the scene
- ▶ 55:46 Jerome Pesenti We're definitely going all in as PyTorch, you know, end to end. 4 times in the scene
A Fresh Approach to Technical Computing // Viral Shah & Stefan Karpinski, Julia Computing
- ▶ 6:39 Stefan Karpinski Um, recently with the advent, like, advent of a lot of these, you know, fairly sophisticated AI frameworks, many of which have, um, you know, like TensorFlow has a Python interface, um, PyTorch is also becoming very popular, um, they have…