Python, every mention
94 scenes (2025) · ← back to Python
tap a year for its mentions
every year 2025 anyone Shawn Wang 47Samuel Colvin 28Chris Lattner 18Bret Taylor 12Jeremy Howard 11Alessio Fanelli 10Gorkem Yurtseven 7Eric Ries 7Akshay Agrawal 7Jack Hopkins 6
Verbatim, from the transcripts: the passages where Python comes up
One Year of MCP — with David Soria Parria and AAIF leads from OpenAI, Goose, Linux Foundation
- ▶ 14:34 David Soria Parra You want to, to have, like, a shared state that you can, like, have, and that's kind of ok, and like, we have seen this in PHP application and Python application being done, but it, it's,
- ▶ 1:00:14 David Soria Parra And so just like being an active part, like either giving us feedback, being in the discord channel, talking with us, giving us ideas, while also just helping us implementing like the types of SDKs, the Python SDKs.
⚡️ 10x AI Engineers with $1m Salaries — Alex Lieberman & Arman Hezarkhani, Tenex
- ▶ 17:59 Shawn Wang Like we think the models are good, but like actually they have been really trained into a certain sort of local minima of like, well, here's all the Python because Sweebench is all Python, all Django.
⚡ Inside GitHub’s AI Revolution: Jared Palmer Reveals Agent HQ & The Future of Coding Agents
- ▶ 27:29 Jared Palmer And like, it's, it ends up doing Python and, and, and
⚡️ The State of AI Engineer Hiring: Cheating, AI Adoption,Junior Devs — Vivek Ravisankar, HackerRank
- ▶ 47:06 unnamed speaker Like I want to change our national second language from Chinese to Python. 2 times in the scene
⚡️Automating Scientific Discovery - Jessica Rumbelow, Leap Labs
- ▶ 12:54 Jessica Rumbelow Maybe like you can get them to use tools and stuff, but still, even if they're, you know, messing around in Python and doing matplotlib and pandas and like, and whatnot, they're still basically
Breaking AI to Fix It: Ian Webster's Journey from Discord's Clyde to Promptfoo's $18M Series A
- ▶ 10:59 Ian Webster But, um, but it can also hook directly into like your Python, your JavaScript, whatever, and we have support for 50 different inference providers.
⚡ Open Model Pretraining Masterclass — Elie Bakouch, HuggingFace SmolLM 3, FineWeb, FinePDF
- ▶ 1:02:12 Shawn Wang Basically, like they're, they're basically offering like a superset of Python that should be good for you, you specifically, you are the target audience for this. 2 times in the scene
Why RL Won — Kyle Corbitt, OpenPipe (acq. CoreWeave)
- ▶ 19:28 Kyle Corbitt If you were to show that exact same equation, just, like, code, not, maybe not PyTorch code, because that you also have to, like, understand, but if you just, like, did the naive implementation in, like, Python, and, like, showed someone,… 3 times in the scene
Building Jamba 3B: the tiny Hybrid Transformer State Space Reasoning Model - Barak Lenz, CTO of AI21
- ▶ 30:33 Barak Lenz Even without the enterprise, you definitely need Python and then web search to, to compensate.
DevDay 2025: Apps SDK, Agent Kit, MCP, Codex and why Prompting is More Important than Ever
- ▶ 25:28 Shawn Wang I thought this was this Python, this is JavaScript, and then there was a whole link in there.
The antidote to AI fatigue — Answer.ai Solveit
- ▶ 9:02 Jeremy Howard Literally any Python function that you put right there in that system is immediately a tool, and it's extraordinary what happens when you write a three-line function that can call grep or rg or whatever, and then you say like, hey, LLM,…
- ▶ 22:37 Alessio Fanelli You know, you might just keep using Solvit and then, you know, you do this for learning Python and then you kind of play out.
- ▶ 40:07 Eric Ries I skipped over this, but in that dialogue, you'll see that it said something like import reader comments as a, as a Python module. 7 times in the scene
Taste is your Moat (Dylan Field of Figma)
- ▶ 3:55 Dylan Field And, uh, it's, I think, very exciting to think about how far you can help people go and how you can make them both more productive, but also help them explore more of the option space of design with some of these techniques.
Amp: The Emperor Has No Clothes
- ▶ 1:08:21 Alessio Fanelli It's like, you know, when, um, Chris Ladner even started working on Mojo, it's like, okay, because of AI, I understand why you need to build a superset of Python.
⚡️ Beyond Transformers with Power Retention
- ▶ 9:30 Diego Bachman That's one of these more higher level Pythonic frameworks.
- ▶ 15:51 Diego Bachman And this is the baseline level of loss that it achieves on like a very standard, they call it the stack data set and specialized to Python.
A Technical History of Generative Media
- ▶ 0:43 Gorkem Yurtseven and then we decided to build a Python runtime in the cloud and that evolved into an inference system that evolved into what Fall is today, which is a generative media platform. 4 times in the scene
- ▶ 5:13 Batuhan Taskaya It's terrible for utilization, and they are not optimizing it, so let's just offer an optimized version of this that's ready for API to be scaled, and doesn't require people to deploy Python code, because we want product engineers to start…
- ▶ 7:54 Gorkem Yurtseven And like, essentially we had this Python runtime. 3 times in the scene
- ▶ 13:42 Batuhan Taskaya I used to work at PyPy, like a just-in-time compiler for Python.
⚡️Launching Ona: Coding Agent with Fully Sandboxed Cloud Environment
- ▶ 11:18 Chris (Christian Weichel) And, you know, for, for weekend warrior where the setting it up on setting up the dev environment might not feel like much, but, you know, if you're working a bit of Python here and a bit of go over here and a bit of typescript over there,…
Greg Brockman on OpenAI's Road to AGI
- ▶ 21:34 Greg Brockman By the way, it does have access to Python tool as well.
⚡️Using RFT to Build Clinical Superintelligence
- ▶ 6:51 Brendan Fortuna Often you could do a unit test, run some Python code, or even use an LLM grader, or even combine a bunch of graders together into a massive grader, you know, kitchen sink.
🕰️ The Oral History of Windsurf (ft. Varun Mohan, Scott Wu, Jeff Wang, Kevin Hou, Anshul R)
- ▶ 1:33:24 Anshul Ramachandran Here is Cascade installing a requirements.txt file and independently realizing that it needed to use Python three instead of Python given my environment.
- ▶ 2:14:30 Shawn Wang I gave, I gave Preb and Kevin a lot of shit for my Python issues.
- ▶ 2:46:38 Kevin Hou And you can see that we're going to be building a Python web scraper. 2 times in the scene
⚡️The Future of Notebooks - with Akshay Agrawal of Marimo
- ▶ 0:48 Akshay Agrawal So Marimo is an open source notebook for Python that's built entirely from scratch. 2 times in the scene
- ▶ 4:38 Akshay Agrawal You have cells of Python code, you have some markdown, and then you have some visual outputs. 2 times in the scene
- ▶ 21:01 unnamed speaker Yeah, I was just kind of curious if there's any part of the Python ecosystem that, you know, is not compatible with this.
- ▶ 24:22 Akshay Agrawal So if I go, uh, let's, I can open a notebook file, and we can take a look at what it actually looks like. 3 times in the scene
Personalized AI Language Education — with Andrew Hsu, Speak
- ▶ 52:41 Andrew Hsu If you've tried coding tools six months ago, and they weren't that great, especially if it's not TypeScript or Python, right?
Information Theory for Language Models: Jack Morris
- ▶ 12:49 Shawn Wang And now he's turned his attention to the Python CUDA relationship, right? 2 times in the scene
⚡️Warp 2.0: the Agentic Development Environment - Zach Lloyd and Ben Holmes
- ▶ 2:55 Shawn Wang And I didn't have some packages and it just, it just started suggesting things to me to fix my Python environment.
- ▶ 21:22 Zach Lloyd So if you're doing like work in Node or Python, Postgres, MySQL, it works in any REPL and it's a, uh, it's a pretty differentiated and cool feature of Warp.
The Shape of Compute (Chris Lattner of Modular)
- ▶ 7:59 Chris Lattner So programming language, syntax, member of the Python family, make it much more accessible and easy to write kernels and performance and all that kind of stuff.
- ▶ 12:01 Chris Lattner It doesn't have Python in the loop and for eager mode op dispatch and stuff like this, because you don't have all that dependency. 14 times in the scene
- ▶ 48:25 Chris Lattner I didn't say it's a replacement for Python. 2 times in the scene
- ▶ 1:09:52 Chris Lattner And also it looks like Python, so you could read it, but it just like massively scales the on-ramp.
Quadratic: The AI Spreadsheet
- ▶ 1:12 David Kirkos So it speaks Python. 2 times in the scene
- ▶ 25:49 unnamed speaker Do you think that will be mostly Python once the model does most of the analysis or what's the kind of like the future of formulas, I guess? 6 times in the scene
The Utility of Interpretability — Emmanuel Amiesen
- ▶ 1:07:43 Emmanuel Ameisen And you, here we do it for like language languages, but I think you could argue that you'd expect the same thing for like programming languages, where it's like, oh, if you learn what an if statement is in Python, maybe it'd be nice if you…
The AI Coding Factory
⚡️Multi-Turn RL for Multi-Hour Agents — with Will Brown, Prime Intellect
- ▶ 33:57 Will Brown Don't, don't have a 2000 line Python script that does that.
DeepWiki: The GitHub Encyclopedia
- ▶ 20:23 unnamed speaker Like I think this part, like when some, if somebody wants to contribute, it's like pretty easy to go through the Python, but then figuring out how everything gets kind of scheduled because it takes so long to do the transcript and all of…
- ▶ 29:43 Silas Alberti And CUDA channels are just so cool because you can just, like, compare the Python implementation with the CUDA implementation. 2 times in the scene
ChatGPT Codex: The Missing Manual
- ▶ 14:04 unnamed speaker Like, Python is one.
Claude Code: Anthropic's CLI Agent
- ▶ 10:42 Boris Cherny It's kind of this research tool that's, uh, you know, it's like written using Python.
- ▶ 37:49 Boris Cherny For other teams, type coverage is very important, especially if you're working in, you know, a very strictly type language and, you know, for example, avoiding like NEs and JavaScript and Python.
Zed Agents — with Zed Cofounders Nathan Sobo & Antonio Scandurra
- ▶ 30:51 Nathan Sobo Well, AI is going to become an ever grow, you know, growing Python part.
Voice AI Masterclass — Kwindla Hultman Kramer and swyx
- ▶ 14:43 Kwindla Hultman Kramer So what we try to do in the pipe cat world is we do a bunch of things in parallel and separate Python tasks and have abstractions for kind of putting those things together or having those things interact in useful ways.
- ▶ 18:35 Kwindla Hultman Kramer It runs Python.
What is an RL environment? w/ Nous Research's Roger Jin
⚡️Factorio Learning Environment: the ultimate Game Agent Eval — Jack Hopkins
- ▶ 5:52 Jack Hopkins And we ultimately settled on a code synthesis approach where these models would write Python code and those Python code in that Python code would invoke certain high level actions, such as move to this distant point or place an entity next… 5 times in the scene
- ▶ 23:23 Jack Hopkins Um, as mentioned, because they execute Python programs, they can kind of come up with quite complicated policies, which a human player would have to kind of develop on the fly.
Why Every Agent needs Open Source Cloud Sandboxes
- ▶ 9:44 Vasek Mlejnsky And, um, what was, I think, especially early on, like Python was the language that the models were probably the best or one of the first languages where the models were working very well. 2 times in the scene
- ▶ 24:08 Alessio Fanelli And then people use depending on the, because the difference that maybe people don't think about is like, you can generate the code with the Python SDK, but you can still run Lua code or like R code. 3 times in the scene
- ▶ 40:50 Vasek Mlejnsky I also like it's like TypeScript first, uh, which I thought like it's the right decision because as, as we talked about, like being more bullish on the product developers, instead of like a Python machine learning type of a developer, I've…
Tiny Teams: $6m ARR, 5m users with 4 employees — Sid Bendre, Oleve (Quizard AI/Unstuck AI)
- ▶ 21:14 unnamed speaker So I see some details of the stack, which is Python, FastAPI, Flask, Mongo, and Postgres.
GPT 4.1: The New OpenAI Workhorse
- ▶ 12:21 unnamed speaker I think I was actually very surprised how all models seem to have more difficulty than I would have expected on a task that, you know, we would find very simple or like, you know, maybe an, uh, an undergrad could write a Python script to,…
Claude Plays Pokémon Hackathon: Escape from Mt. Moon!
- ▶ 22:52 David Hershey So it's actually still just, like, represented as a Python dictionary, but it can choose to, like, load or unload a file from its memory.
- ▶ 29:50 unnamed speaker And Python's really not that bad anymore with UV.
- ▶ 1:04:16 Jesse Han Um, so specifically, what you'll need to submit is that you'll need to submit a GitHub gist that implements a subclass of the Eva agent, um, and, uh, you can use this, uh, Python file as a template, uh, and however you modify it, it needs… 2 times in the scene
The Creators of Model Context Protocol
- ▶ 10:48 unnamed speaker Um, that was TypeScript, Python, and then for the Zed integration, it was Rust.
The Agent Network — Dharmesh Shah, Agent.ai + CTO of HubSpot
- ▶ 26:14 Dharmesh Shah So I have a 14 year old right now and is taking Python programming class.
- ▶ 49:12 Dharmesh Shah The fact that it has access to a Python interpreter that I can actually call at runtime, that solves an entire body of problems that it wasn't trained to do. 4 times in the scene
Fullstack-Bench: The Eval for Coding Agents — with Sujay Jayakar, Chief Scientist, Convex
- ▶ 3:45 Sujay Jayakar Being able to now reproduce this and see how changes to the platform or changes to the backend, you know, like we, this framework supports convex, but we also compare it to Supabase and we compare it to like a more traditional kind of…
Snipd: The AI Podcast App for Learning — with CEO Kevin Ben-Smith
- ▶ 28:39 Kevin Ben-Smith So the general tech stack is our entire back end is, or 90% of our back end is written in Python.
- ▶ 53:45 Shawn Wang Uh, is it Django, just like the Python one?
[Lightning Pod] Evals: How to Improve AI Consistently — with Hamel Husain and Shreya Shankar
- ▶ 21:03 Shawn Wang I don't know if you guys have seen, uh, you know, in the way, in the same way that I think, uh, Hamill likes Marimo, which is like a sort of Python reinvention of the notebook. 2 times in the scene
Solve coding, solve AGI [Reflection.ai launch w/ CEO Misha Laskin]
- ▶ 12:06 Shawn Wang Yeah, so, like, and then I think one more thing that, like, ties to a previous conversation we had with, uh, Brett Taylor from OpenAI, which was, like, a lot of our code right now is just in languages that happen to be nice for humans to… 2 times in the scene
How Claude Plays Pokémon was made
- ▶ 10:44 David Hershey Cause I could have done it, but there's a lot of like tedious here are addresses in memory, map that to a Python program that I had no interest.
- ▶ 26:53 David Hershey Like, it's, like, more or less a Python dictionary that's appended to the prompt.
smol agents are all you need
- ▶ 7:21 Aymeric (Emmerich) I thought that this had potential because they had built from the ground up a new Python interpreter, uh, basically making code agents, um, has big security, um, challenges. 3 times in the scene
The AI Architect: Bret Taylor
- ▶ 38:54 Bret Taylor So the fact that we're still writing Python with AI cracks me up just cause it's like literally was designed to be ergonomic to write, not, not safe to run or fast to run. 12 times in the scene
Agent Engineering with Pydantic + Graphs — with Samuel Colvin, CEO of Pydantic Logfire
- ▶ 1:46 Samuel Colvin I think the success of Pydantic and libraries like fast API that build on top of it means that today that's no longer controversial in Python.
- ▶ 3:58 Samuel Colvin Not convert to Python types. 2 times in the scene
- ▶ 5:52 Samuel Colvin Like we're going to go off and three engineers are going to work full on for a year in Python and Rust. 3 times in the scene
- ▶ 8:11 Samuel Colvin Looking in general at the ecosystem of agent frameworks, the engineering quality is far below that of the rest of the Python ecosystem. 4 times in the scene
- ▶ 11:11 Samuel Colvin I was sort of of the opinion you didn't need them, and you could use standard, like, Python flow control to do all of that stuff. 2 times in the scene
- ▶ 14:32 Shawn Wang Like, um, it looks like normal Pythonic code, but you just have to keep in mind what the type-ins actually do with the quote-unquote magic that the graph construction does. 5 times in the scene
- ▶ 21:48 Samuel Colvin Because my, like, just call the next agent with Python code was not a satisfactory answer to people, so it was like, okay, we gotta go and have a better answer for that, that, like,
- ▶ 32:02 Alessio Fanelli I think you have busy like this test model where like just through Python, you try and figure out, uh, what the model might respond without actually calling the model.
- ▶ 50:11 Samuel Colvin If you search Datadog logging Python and you just try as a like a non-observability expert to get something up and running with Datadog and Python. 4 times in the scene
The Agent Reasoning Interface: Claude, ChatGPT Canvas, Tasks, Operator — with Karina Nguyen, OpenAI
- ▶ 40:45 Karina Nguyen But it's the same as in decision boundary between like Python, ADA advanced data analysis versus Canvas is one of the most trickiest like decision boundary behaviors that we had to like figure out.
- ▶ 43:09 Karina Nguyen But then the magic is like when I actually, because the model is so general, it knows how to use search or like canvas or like create sci-fi stories and create Python puzzles.
Beating OpenAI and Anthropic by Looking At Data: the new #1 on SWE-Bench w/ W&B CTO Shawn Lewis
- ▶ 21:12 Shawn Lewis So the agent actually sees every line in this Python file prefixed with line numbers that look like this.
Outlasting Noam Shazeer, Crowdsourcing Chai AI w/ 1.4m DAU — with William Beauchamp, Chai Research
- ▶ 2:49 William Beauchamp So started off and the taught myself Python,
- ▶ 12:07 William Beauchamp But anyone could write a Python script, they could submit it to the CHI backend, and we would then host this code and execute it. 4 times in the scene
AI Engineering for Art - with comfyanonymous
- ▶ 2:09 Shawn Wang but like, already some interest in automations, probably a lot of Python. 2 times in the scene
- ▶ 28:50 Shawn Wang I feel like if Streamlit or Gradio offered something that you would have used Streamlit or Gradio because it's Python. 3 times in the scene
- ▶ 33:17 comfyanonymous (Comfy) Don't have that much fine-grained control over, like, uh, specific memory stuff, so, uh, kind of have to leave, like, the memory freeing to, to Python and PyTorch, which is, can be annoying sometimes.
The State of Reasoning — from Nathan Lambert, Interconnects/AI2 [LS Live @ NeurIPS 2024]
- ▶ 11:57 Nathan Lambert This on the screen is an example for precise instruction following, which is the idea that if you have a prompt, you can say something like, um, have every sentence start with the letter a, and you can verify that with Python really easily. 2 times in the scene