ATRIUMsearch → argument graph
Video · 2026-08-12 · 60m · 6 moments

The Workflow of the Future With Zed

✦ AI generated

timeline · colored by role

01
Anecdote

Pair programming is a transformative learning experience that improves software development skills rapidly.

Nathan Sobo describes his experience at Pivotal Labs with pair programming, where working closely with another developer significantly accelerated his learning and led to the vision for Zed to replicate this collaborative experience over the internet.

transcript

Nathan Sobo: And that taught me so much. I got I became such a better software developer so rapidly. Um but I always wanted a tool that could replicate that experience that we had in this shared office, like over the internet, and turn the internet into that whole experience. And Zed is as close as I've ever gotten um until what comes next.

02
Mechanism

Using web technology (Electron) for desktop apps limits performance and user experience, necessitating a move to native code with Rust and a custom UI framework.

After creating Atom with Electron, Nathan Sobo realized the browser abstraction overhead limited performance and aesthetics, leading to the decision to build Zed in Rust with a custom UI framework (GPUI) for direct hardware control.

transcript

Nathan Sobo: the trade-off that we made, the devil's bargain that we made was that um the overhead of the browser and sort of the abstraction that it presents, which is this kind of lowest common denominator common standard that needs to run in a backward compatible way on all these different browsers. Well, all of that is reducing down, but still it's like it really put a a limit on the performance of and the user experience that we could yield. And so, after Atom sort of ran its course and Microsoft acquired GitHub and uh I was sitting there using Visual Studio Code, I just wasn't happy. Like I just couldn't see a future where I continued writing all my software in this web browser masquerading as a desktop app. Um just what the performance wasn't there, the aesthetics weren't there for me, this collaborative dimension even though they did introduce live share, like it never really worked as well as I wanted it to. Um and so that led me to one inexorable conclusion that we had to get closer to the metal.

03
Example

AI agents are used extensively in the development workflow, especially for code review and debugging, but with human engagement to ensure quality.

Nathan Sobo describes how the Zed team uses AI agents extensively in their workflow, particularly for interpreting diffs during code reviews and getting a head start on stack traces, while maintaining human engagement to ensure quality.

transcript

Nathan Sobo: we use AI a lot. Um at least I do, and I think it just it varies across the team and obviously most of the well, everybody on the team is an engineer back from the days when you didn't have an agent to write any code for you. You were typing it out and you know, you were lucky if you could get a auto complete. Um so, I'm really glad that I have that pre-training of my neurons in place to really understand software. Like I I think it would be a different story if I were getting my start now. Some things would be easier and then certain kinds of lessons might be harder. Um but I I write a ton of code agentically these days. Um and I used it in the pull request review process. When I'm reviewing pull requests, it's amazing to get some help interpreting what the heck is going on in this diff.

extends · 1gives example · 1supports · 1

04
Claim

A version control system that tracks every change connected to the conversation that shaped it enables continuous, real-time collaboration with agents and teammates.

Nathan Sobo introduces Delta DB, a version control system that records work as it unfolds, keeping every change connected to the conversation, enabling conflict-free work trees and a new way to collaborate with agents and teams in real-time.

transcript

Nathan Sobo: And so DeltaDB has been our vision for yeah, giving basically every version of the software as it evolves an address uh that you can recall it, talk about it, and you can drape over that level of tracking the ability to comment on any part of the code at any time, and thus be able to have a conversation sort of in the code base as you're working on it.

extends · 1gives example · 1provides context · 1

05
Claim

Conversations are becoming the source of code, with code as a downstream artifact.

Nathan Sobo argues that with the rise of AI agents, conversations now initiate the coding process, making code a product of those conversations rather than the starting point.

transcript

Nathan Sobo: now conversations are becoming sort of the source of our code. Like we were thinking we'd write all the code and have conversations in it. Now conversations are the beginning and code comes out of that. It's a downstream artifact.

06
Prediction

The ultimate user interface is language, and as natural language becomes more accessible, it will replace many gestural user interfaces.

Nathan Sobo predicts that language will become the primary user interface, reducing the need for complex graphical interfaces, though graphics will still have a place for appropriate contexts.

transcript

Nathan Sobo: the ultimate user interface is basically language. Um and you know, for the longest time for the last whatever 50 years, 40 years, like the only people that could really use language to talk to a computer were people like us, software developers. And uh even then it was like pretty grueling brutal process of memorizing a menagerie of commands and flags and programming languages and tags and all of it, right? Um so, I think now that language natural language is like opened up as a way of interacting, like it's such a superior means of expression over essentially like gestural user interfaces that we've been confined to use in the previous era.

extends · 1provides context · 1supports · 1

Highlight slides
Related episodes