2024
Lex Fridman said he recently moved from the VS Code editor to Cursor, an LLM-based editor, having earlier moved from Emacs to VS Code.
“I recently switched from VS Code to Cursor, and before that it was Emacs to VS Code switch.”
Topic
Cursor
42 facts from 22 episodes, in date order. These are what speakers said on the shows, each linked to the moment it was said. They aren't independently verified.
2024
Dohmke says Cursor was first to adopt Claude Sonnet 3.5 and to offer developers a choice of multiple models.
“They were the first, you know, with Claude Sonnet 3.5, so they were also, you know, ahead of the curve of saying, okay, multiple models”
2025
Praveen Akkiraju says Cursor seems by far the most popular AI coding tool among developers.
“I mean, Cursor, by far, seems to be the most popular among developers today”
2025
Pavel Veller said Cursor had recently introduced a Max mode billed per use, which runs more iterations and spends more tokens.
“Uh, they've introduced very recently cursor on on Sonnet 3.7 and now on Gemini 2.0. I think they have a max mode, which is pay per use”
2025
Dohmke says GitHub believes Copilot's business is bigger than Cursor's.
“given that, you know, we are very open about that, we believe Copilot is, uh, bigger than Cursor”
2025
Stack Overflow launched MCP servers for its public platform and enterprise product so tools like Cursor can read from, and write back to, its content.
“We've launched, for example, MCP servers for both our public platform as well as our enterprise product.”
2026
Zapier CEO Wade Foster says he mostly uses the Cursor AI coding tool as his daily driver.
“Yeah, so I mostly use cursor every single day.”
2026
Angela Jiang says Anthropic runs early access programs with API customers such as Cognition and Cursor before model launches.
“we've engaged with, uh, a lot of customers, like, like Cognition, like, like Cursor”
Undated
Bret Taylor cites Microsoft Copilot and Cursor, which his company uses, as examples of AI agents that help software engineers code.
“Microsoft Copilot, Cursor, which is a cool startup that my company uses.”
Undated
Aaron Levie says coding assistants like Copilot and Cursor took off because they add productivity to an existing workflow without requiring users to change behavior.
“I would say this largely explains the breakthrough in copilot or cursor I is because I get to do exactly the same thing I'm doing now, but I'm seeing incremental productivity gains right away without changing really any behavior.”
Undated
Praveen Akkiraju says Devin works differently from Cursor by firing off a bunch of agents that execute the user's plan.
“If you take something like Devin has a different sort of way it it engages, you know, it fires off a bunch of agents that, you know, execute your plan”
Undated
Moura says Cursor supports custom per-project rules that are embedded into prompts, and that his team uses them.
“is that you can add custom cursor rules, and I think people don't realize that, and you can do that per project.”
Undated
Sualeh Asif says Cursor co-founder Michael had spent considerable time building high-performance games.
“I think Michael had spent quite a bit of time building these, uh, high performance games”
Undated
Sualeh Asif says the Cursor founders did competitive math in college, not only competitive coding.
“I think we're not only competitive coders, like we did competitive math in college.”
Undated
Cursor built about three agent prototypes before shipping its agent, because the models would lose track of the task.
“and we had probably done three agent prototypes before that. We didn't ship because some version of the the model would just lose track”
Undated
Sualeh Asif says Cursor's custom tab model handled about 100 million requests per day and was growing quickly.
“For example, for our custom tab model, we do something like 100 million requests a day, uh, and and and quickly growing it”
Undated
Cursor trains a large first version of its apply model, deploys it to users, and uses the resulting data to distill smaller, faster models.
“you train your 1st version of an apply that is quite a bit bigger. And, uh, you then deploy it for all users. You get lots and lots of data and then you can distill a slightly smaller model”
Undated
Sualeh Asif says Cursor's apply feature is a model rewriting the entire file rather than a deterministic algorithm, and that improvements came largely from making the model smaller.
“It's it's a model that is actually rewriting the entire file and a lot of, you know, a lot of the improvements have been making the model smaller.”
Undated
Sualeh Asif says early language models used context windows of about 4,000 to 8,000 tokens, which limited what a coding agent could do.
“when the models started out, you were sort of doing these 4k, 8k context windows.”
Undated
Inside Cursor's code base, the Cursor Tab feature is named CPP, short for 'copilot plus plus'.
“cursor tab in our code base is named CPP for being copilot plus plus”
Undated
Sualeh Asif says Cursor tries never to store user code beyond the life of a request, though it hopes to store some code privately in future.
“we try to make sure that the that we never store any code past the life of your request.”
Undated
Cursor's code-indexing infrastructure runs at a scale of billions of files per day, using Cursor's own inference for embedding models.
“the one that we've been sort of very public about is our indexing infrastructure that, uh, we spent a lot of time optimizing and running at quite enormous scales, like billions of files, uh, per day kind of infrastructure”
Undated
Cursor is working on an index architecture with a shared 'trunk' of a code base plus per-user branches, to avoid thousands of duplicate copies.
“can we have a shared trunk? And then, uh, every single person can have their sort of branch off that trunk?”
Undated
Sualeh Asif says Cursor has a fixed number of GPUs at any moment, which limits its token capacity and forces trade-offs in indexing.
“we at any point in time have a fixed number of GPUs, which means we have a fixed num amount of token capacity.”
Undated
Indexing uses a small share of Cursor's GPUs, with the tab model and larger models using far more compute.
“we run the tab model, like indexing is a very small percentage of, uh, of our GPUs.”
Undated
Sualeh Asif says Cursor's tab model serves hundreds of millions of calls per day.
“so tab model is like hundreds of millions of calls per day”
Undated
Sualeh Asif says Cursor chose DeepSeek models as a pre-trained base for its own post-training and reinforcement learning stack.
“for just picking a really well pre trained base, uh, DeepSeek does a phenomenal, phenomenal job.”
Undated
Sualeh Asif says reinforcement learning and post-training have made models much more coherent over tens of thousands of output tokens, which helps agents.
“one of the things that is talked about less is how coherent the models have gotten over producing tens of thousands of tokens of output”
Undated
Cursor uses an internal prompt-building library, called Preempt in the transcript, built to suit its own needs.
“we use this internal library called Preempt Um, and the way we built Preempt was it was well suited to our own needs at the time.”
Undated
Sualeh Asif says Cursor's evaluation of its agents was largely vibes-based at the time, though he expects more daily metrics later.
“it's, it's pretty, pretty vibes based. I suspect it'll be vibes based in the short term”
Undated
Aaron Levie says Cursor is the fastest growing software company in history, at least in an enterprise context.
“As far as every stat I've seen, Cursor is the fastest growing software company in history, at least in an enterprise context.”
Undated
Aaron Levie says that, by every statistic he has seen, Cursor is the fastest-growing software company in history, at least in an enterprise context.
“As far as every stat I've seen, cursor is the fastest growing software company in history, at least in an enterprise context”
Undated
Josh Weiner says he has been building an AI chess coach using tools such as Cursor and GitHub Copilot.
“I've been in the process of building out an AI chess coach using some of these tools like Cursor or GitHub Copilot and so forth.”
Undated
Kenny Song says he enjoys OpenAI's o3 model in ChatGPT and the Cursor tool, and expects tools to become more agentic.
“I really loved O3 as a model in in ChatGpt, and I love using cursor, and I'm excited for these tools to become more and more agentic over time.”
Undated
Elad Gil said Cursor launched its IDE roughly two years before the recording.
“Cursor didn't really launch its IDE until 24 months ago, something like that, probably.”
Undated
Gultekin cites Cursor as an example of a company for which a custom autocomplete model makes sense at its scale.
“At their scale, it does make a lot of sense for them to have a custom model that is doing their, you know, autocomplete, for instance.”
Undated
Tsavo Knott argues companies need one universal context layer that any AI tool, such as Cursor, Codex or Claude, can draw on.
“you need one universal context substrate across everything that you're doing that, you know, independent of what you're using, be it cursor, codex, cloud, independent of of any system you're using, they can tap into it.”
Undated
Tsavo Knott used sub-agents in Cursor to review his previous twelve months of projects and time spent, in order to prepare Pieces's research and development tax credit.
“I just in cursor, uh, used a variety of sub agents to say, Hey, look at my last twelve months. What are all the projects that I worked on? How much time did I spend on them?”
Undated
Ed Sim says Cursor first post-trained a Chinese open-weight model and later built its own model, using data generated by its users.
“They post-trained their own model initially using a Chinese open weight model, then they built their own.”
Undated
Datadog began its coding-agent rollout with roughly 50 Cursor licenses.
“I remember our 1st install, we started with Cursor, and I remember getting like 50 licenses.”
Undated
At VMware Explorer, nearly every attendee raised a hand when asked if they used tools like Cursor or Claude Code at work.
“And pretty much the entire room, and all of them went up, and I was like, Well, that's a lot more than I expected.”
Undated
Zach Lloyd says Cursor made a good exit by becoming a model lab after recognizing the economics of building on third-party models.
“I mean, Cursor was brilliant, they did it, and then I think that they had a perfect exit and became a model lab because I think they fully realized the economics of it.”
Episodes
- 3 Sep 2024#442 – Donald Trump Interview1h 12m
- 3 Nov 2026⚡️ The 2026 Software Factory Standard — Zach Lloyd, Warp.dev34m
- 27 Oct 2026How Anthropic ships the #1 LLM in the World safely: Separating the Brain from the Hands1h 46m
- 27 May 2025Delivering More Connected Care Through AI: CVS Health's Josh Weiner22m
- 10 Aug 2026Enterprise AI Biggest Opportunities: A Top VC's Take55m
- 12 Feb 2025An Advanced Look at AI Agents, Part 2 | #86853m
- 15 Dec 2025Stack Overflow users don't trust AI. They're using it anyway1h 11m
- 7 Aug 2025GitHub's CEO says AI coding is ‘here to stay’58m
- 24 May 2025Rapid Response: The enterprise AI revolution is underway, with Box’s Aaron Levie36m
- 20 Nov 2024How agentic AI will save customer service, with Bret Taylor38m
- 5 Dec 2025Scaling Legal AI and Building Next-Generation Law Firms with Harvey Co-Founder and President Gabe Pereyra44m
- 20 May 2025The enterprise AI revolution is underway, with Box’s Aaron Levie40m
- 11 Aug 2026The CISO Playbook for AI Agents | Datadog22m
- 29 Apr 2025Inside Cursor: The future of AI coding with Co-founder Sualeh Asif49m
- 25 Feb 2025The rise of AI agents49m
- 9 Jul 2026How AI Is Transforming Governance and Workflow Automation in the Enterprise - with Tsavo Knott of Pieces41m
- 17 Sep 2026No Code Is Code: Zapier CEO Wade Foster on Headless Tools, Zapier MCP & Automation Bench1h 6m
- 14 Jan 2026Snowflake VP of AI Baris Gultekin on Bringing AI to Data, Agent Design, Text-2-SQL, RAG & More1h 36m
- 29 Jan 2025Aaron Levie, CEO of Box, on Box AI, Enterprise Enthusiasm, and the Evolution of SaaS50m
- 24 Sep 2026From AGENTS.md to Enterprise Deployment48m
- 27 Aug 2025GenAI risks and global adoption43m
- 14 Apr 2025Orchestrating agents, APIs, and MCP servers42m