Andrej Karpathy, every mention
69 scenes · ← back to Andrej Karpathy
tap a year for its mentions
every year anyone Shawn Wang 18Mikhail Parakhin 6Alessio Fanelli 5Raiza Martin 4Lance Martin 4Gavriel Cohen 4Jeremy Howard 3Eugene Cheah 3Dex Horthy 3Suhail Doshi 2
Verbatim, from the transcripts: the passages where Andrej Karpathy comes up
The AI Frontier: from open weights to open research — Eiso Kant, Poolside AI
- ▶ 2:41 Eiso Kant Um, I owe getting into this space to Andre Karpathy.
- ▶ 51:15 unnamed speaker Another Andre invention, um, coinage, uh, which is like, I honestly, like how many objective functions can there be, right?
- ▶ 1:36:15 Eiso Kant I know that people like, it's easy to give them flack, but I also want to say, like, I remember when we started Sourced, right, in, in, in, in, in a, in a, in a, in a, in a post that Kapathi article, it was able for this progress to happen…
- ▶ 1:43:10 unnamed speaker But, you know, part of Karpathy's thesis was cognitive core, right?
Podcast Crossover: AIE, AGI, frontier lab strategy with @matthew_berman and @swyxtv
- ▶ 0:48 Shawn Wang Um, and so I bought the domain AI.Engineer and planned this conference, and I wrote the blog post, and it really kicked off when Andre Karpathy also endorsed it.
The Future of AI Infra: from Kubernetes to Agent Sandboxes — Akshat Bubna, Modal CTO
- ▶ 35:13 unnamed speaker So my impression is, auto research means many things, like anything that Andre coins.
The Blueprint for Autonomous Work Agents | Gavriel Cohen, NanoClaw
- ▶ 1:27 Gavriel Cohen And he's got this kind of second brain type of setup where he's, uh, got a Karpathy LLM wiki, uh, with embeddings and some memory system.
- ▶ 11:32 Gavriel Cohen I just point my agents at Karpathy's, uh, posts, and I say, read this.
- ▶ 14:55 Gavriel Cohen And then Karpathy, uh, tweeted about it and it went to just a whole nother level. 2 times in the scene
⚡️ Google's Open AI Strategy — Omar Sanseviero, Google DeepMind
- ▶ 24:19 Omar Sanseviero I don't know, like, with Carpathic experiments, it's been quite interesting to see, like, how things are evolving. 2 times in the scene
🔬How GPT‑5 derived new results in theoretical physics and quantum gravity — Alex Lupsasca, OpenAI
- ▶ 3:54 unnamed speaker Starting sometime last fall, especially, uh, it just took off, and a bunch of people are like, even like Andre Karpathy went from,
AI-Native Engineering: 100% adoption, 5x search throughput, unlimited tokens — Mikhail Parakhin
- ▶ 3:48 Mikhail Parakhin We just, uh, just even yesterday, Andrej Karpathy was famous in tweeting about, oh, there's some, uh, ways that you can organize your agents to store the data and then look up the data so that you don't have to research or lose context…
- ▶ 28:30 Mikhail Parakhin Uh, you know, effectively, effectively, Andrej Karpathy recently popularized it with Autoresearch. 5 times in the scene
Extreme Harness Engineering: 1M LOC, 1B toks/day, 0% human code or review — Ryan Lopopolo, OpenAI
- ▶ 46:51 unnamed speaker Like, you know, like, way back at the start of this whole AI wave, like, uh, Andre was like, you know, English is the hottest new programming language.
Moonlake: Interactive, Multimodal World Models — with Chris Manning and Fan-yun Sun
- ▶ 1:00:12 Chris Manning And, you know, there were, there were links across at that stage as well, so, I mean, you know, there were several papers in that era of doing, I mean, so Andre Kapathi was a PhD student at the same time as Richard, and so there was some…
Why Your AI Agents Don’t Work with Dex Horthy of HumanLayer | In-Context Cooking
- ▶ 3:42 Dex Horthy According to SWIX, there are, there are many people who, who, uh, Gemini thinks it was Andre Karpathy. 3 times in the scene
Cursor's Third Era: Cloud Agents — ft. Sam Whitmore, Jonas Nelle, Cursor
Captaining IMO Gold, Deep Think, On-Policy RL, Feeling the AGI in Singapore — Yi Tay
- ▶ 1:29:45 unnamed speaker Kapati-inspired, like, biohacking?
Steve Yegge's Vibe Coding Manifesto: Why Claude Code Isn't It & What Comes After the IDE
- ▶ 16:20 Steve Yegge That Jeffrey Emanuel, the way that an incredibly well-written article that said, this is why it's a bubble, and the whole market went, and Karpathy started following.
AI to AE's: Grit, Glean, and Kleiner Perkins' next Enterprise AI hit — Joubin Mirzadegan, Roadrunner
- ▶ 1:08:43 Shawn Wang Is a supporter of ours that is, that has promised to be on at some point, uh, Andre. 2 times in the scene
World Models & General Intuition: Khosla's largest bet since LLMs & OpenAI
- ▶ 1:03:02 Pim de Witte Um, the sequences of AI in three stages, and I credit Andre Kaparthi for, for teaching this, um, bits to bits, atoms to bits and bits to atoms, and then atoms to atoms.
After LLMs: Spatial Intelligence and World Models — Fei-Fei Li & Justin Johnson, World Labs
- ▶ 13:36 unnamed speaker Coming back to the little bit of background in history, we have this sort of research note on the scene storytelling work that you did, or neural, neural image captioning, uh, that you did with Andre. 10 times in the scene
⚡️ 10x AI Engineers with $1m Salaries — Alex Lieberman & Arman Hezarkhani, Tenex
- ▶ 18:35 Arman Hezarkhani I think that it's context engineering in Andre Karpathy's words, right?
- ▶ 22:28 Alex Lieberman Then he's like, if you want to go super deep, do any of Andre Karpathy's 2 times in the scene
⚡ Inside GitHub’s AI Revolution: Jared Palmer Reveals Agent HQ & The Future of Coding Agents
- ▶ 34:59 unnamed speaker Yeah, I think, like, Audrey Karpathy was, the other day, was saying, like, every company needs one of these, where you can just, like, hey, like, this, this really should exist at GitHub.
⚡ Inside Google Labs: Building The Gemini Coding Agent — Jed Borovik, Jules + AIE CODE Preview
- ▶ 36:38 unnamed speaker Andre and he's super influential in February and like people have just come to kind of use it as a blank check to just YOLO on prompts and stuff and then create the worst code imaginable and leave other people to clean it up.
How Zyphra went all-in on AMD + Why Devs feel faster with AI but are slower — with Quentin Anthony
- ▶ 48:51 Quentin Anthony I think Carpathia made a good, uh, tweet about like physicists are like the embryonic stem cells, right?
The Agents Economy Backbone - with Emily Glassberg Sands, Head of Data & AI at Stripe
- ▶ 28:22 Emily Glassberg Sands Did you guys see Karpathy's tweet about the, he basically like recreated
The antidote to AI fatigue — Answer.ai Solveit
- ▶ 46:08 Jeremy Howard And so, we were like, this is such a great way to turn, um, discussions into, into text, that we took on Andre Capathy's challenge, and his challenge was, hey, can anybody take my, let's build the GPT tokenizer video, and turn it into a…
⚡️ Beyond Transformers with Power Retention
- ▶ 21:12 unnamed speaker And now, um, if we go into model.py, you can see this is just like a very ordinary, just Karpathy style setup.
⚡️No, Don't Do Palantir for AI - Brendan Falk, Hercules (AUDIO FIXED)
- ▶ 29:55 Shawn Wang I think, uh, Karpathi talked about it when he was talking about his Menugen app for people who don't, uh, know they can look that up.
Context Engineering for Agents - Lance Martin, LangChain
- ▶ 1:31 Lance Martin Karpathy put out that tweet kind of canonizing the term context engineering, and he kind of mentioned this, uh, kind of nice definition, which is context engineering is the challenge of feeding an LM just the right context for the next…
- ▶ 1:02:14 Lance Martin So I, I actually was very inspired by, uh, Karpathy had a tweet a long time ago talking about building on ramps. 3 times in the scene
Better Data is All You Need — Ari Morcos, Datology
- ▶ 1:05:07 Shawn Wang I think there's, there's another version of this, which is the sort of Andre Karpathy cognitive core concept of a model that doesn't know anything, but can use tools a lot to, to find, to figure out.
Personalized AI Language Education — with Andrew Hsu, Speak
- ▶ 2:59 Andrew Hsu We talked to Karpathy back then, actually, when he was, like, just finishing grad school, and did a lot of sort of self-study research, and we were just so convinced, I think fundamentally, that speech models were going like this, language…
- ▶ 1:00:50 Shawn Wang One on Andre, you talked to him in your machine learning journey. 6 times in the scene
⚡️Anthropic vs Cognition on Multi-Agents: A Breakdown with Dylan Davis
- ▶ 13:22 Shawn Wang Everyone's talking about it, uh, only cause Andre talked about it, but also like it was already trending before, before Andre talked about it.
Information Theory for Language Models: Jack Morris
- ▶ 40:35 Shawn Wang I think maybe this is a good point to maybe also bring in the idea that Andre has been pushing for the last like, I think year and a bit of, um, the cognitive core, like what is the dumbest possible model that knows nothing, but is, is… 3 times in the scene
- ▶ 55:48 Shawn Wang Because like, that's ultimately how we figure out this whole compression issue and the, you know, what, what Andre wants, which is like the cognitive core, right?
⚡️Launching AI Diplomacy: the hardest LLM Game Benchmark yet - Alex Duffy
- ▶ 7:26 Alex Duffy Karpathy was like, games are a really great area to, to explore.
- ▶ 15:02 Alex Duffy Um, you can, yeah, in my pin post, like one of my first responses to it was, if you scroll down, I like respond, Andre Carpathie, yeah, the bottom video.
GPT 4.1: The New OpenAI Workhorse
- ▶ 6:48 Shawn Wang We are, our discord actually also did a launch, uh, watch party for the recent, uh, 4.5 podcasts that Sam Altman did, where I think for the first time, it was basically kind of confirmed that like something that people already knew, like…
Snipd: The AI Podcast App for Learning — with CEO Kevin Ben-Smith
- ▶ 40:40 Kevin Ben-Smith This is like, you don't have to be like, I don't know, Andre Kapathi to come up with that and build that, right?
Browserbase: Browser Infrastructure For Your AI Agents
- ▶ 10:18 Paul Klein I guess we go back to the LLMOS talk, uh, that Karpathy gave way long ago, and like the browser box was very clearly there, and it seemed like the people who were building in this space also agreed that browsers are a core primitive of…
- ▶ 36:28 unnamed speaker But it's just an observation that like I, I, in the same way that I see each option, each checkbox in the side of custom GPT is becoming a startup or each box in the Karpathy chart being a startup, like this is also becoming a thing.
The AI Architect: Bret Taylor
- ▶ 1:09:33 Bret Taylor I think it was Andrei Karpathy, but I can't remember, but
Beating Google at Search with Neural PageRank and $5M of H200s — with Will Bryk of Exa.ai
2024 in Post-Transformer Architectures: State Space Models, RWKV [Latent Space LIVE! @ NeurIPS 2024]
- ▶ 21:19 Eugene Cheah Uh, Kapati is quoted saying that RNNs are actually unreasonably effective.
How NotebookLM Was Made
- ▶ 29:38 Raiza Martin Um, I think that's what Karpathy did, right? 4 times in the scene
[Paper Club] Molmo + Pixmo + Whisper 3 Turbo - with Vibhu Sapra, Nathan Lambert, Amgadoz
- ▶ 1:09:22 unnamed speaker So one of them is like the GPU mode LLM.c taught by Karpathy. 3 times in the scene
[LLM Paper Club] Llama 3.1 Paper: The Llama Family of Models
- ▶ 14:05 unnamed speaker Other tweets, George hots Carpathy tweeted, um, VLL supports it.
- ▶ 32:33 Eugene Yan They, they could not have done it without that, and Andre Kapati had a great tweet about this, whereby 2 times in the scene
Training Llama 2, 3 & 4: The Path to Open Source AGI — with Thomas Scialom of Meta AI
- ▶ 49:33 Alessio Fanelli This is kind of like Undergarpati's idea of like more tokens equal more thinking.
The 10,000x Yolo Researcher Metagame — with Yi Tay of Reka
- ▶ 1:25:51 unnamed speaker Uh, I do think Andre Kapathy mentioned that sometimes you, like, the best researchers do their own evals.
- ▶ 2:04:00 unnamed speaker I'm looking for, like, a practice where, like, okay, uh, so Andre recently had a thing where, like, before, when he wakes up, he doesn't look at social media, he only goes straight to cook, straight to work. 2 times in the scene
High Agency Pydantic over VC Backed Frameworks — with Jason Liu of Instructor
Making Transformers Sing - with Mikey Shulman of Suno
- ▶ 16:46 Mikey Shulman Uh, there's a big shout out to, um, Andre Karpathy's nano GPT.
- ▶ 49:15 Alessio Fanelli I think Carpathi also tweeted something. 4 times in the scene
A Brief History of the Open Source AI Hacker - with Ben Firshman of Replicate
- ▶ 12:07 Ben Firshman We're both users of Archive Sanity, which is like a sort of archive, which is Andre's, uh, like Rex is on top of Archive.
The Four Wars of the AI Stack - Dec 2023 Recap
- ▶ 19:14 unnamed speaker Andre Karpathy, um, at NeurIPS highlighted this paper from DeepMind where, um, they were, they were bootstrapping synthetic data that could be verifiably, uh, proven correct.
The AI-First Graphics Editor - with Suhail Doshi of Playground AI
- ▶ 7:23 unnamed speaker Like, I think, uh, Andre Kapathi has always been, has been making a lot of analogies with the LLMOS. 2 times in the scene
- ▶ 1:07:34 Suhail Doshi You should just go to YouTube and visit Kaparthi's, uh, class. 2 times in the scene
Beating GPT-4 with Open Source Models - with Michael Royzen of Phind
- ▶ 1:10:51 Michael Royzen The product, the UI, the UX is kind of model agnostic, but when it has a better kind of kernel, as Andre Karpathy put it, plugged into it, um, it gets so much better.
Why AI Agents Don't Work (yet) - with Kanjun Qiu of Imbue
- ▶ 47:06 Shawn Wang Um, I think, I think Andre Kapathi has already made this comparison, so there's nothing new here.
The End of Finetuning — with Jeremy Howard of Fast.ai
- ▶ 12:21 Jeremy Howard And in fact, uh, Andre Capathy did his fantastic, uh, RNN demonstration from this at a similar time where he showed, like, you can have it ingest Shakespeare and it will generate something that
- ▶ 28:15 Jeremy Howard Yeah, actually, Andre Capathy grabbed me when I saw him at NeurIpes a few years ago, and he's like, I have to tell you, thanks to the fast AI courses, when people come to Tesla, and they need to know more about deep learning, we always…
RWKV: Reinventing RNNs for the Transformer Era
- ▶ 1:39:15 Eugene Cheah Karbati. 2 times in the scene
Ep 18: Petaflops to the People — with George Hotz of tinycorp
- ▶ 54:05 George Hotz I just, like, you know, wrap my head around, I couldn't, like, like, wrap my head around this, and this was around the time Carpathia released the first, like, uh, RNN that generated the Shakespeare stuff, and I'm like, okay, I get it,…