Scott Wu
How Devin replaces your junior engineers with infinite AI interns that never sleep, covering AI product work, product design, and engineering tradeoffs.
Episode
How Devin replaces your junior engineers with infinite AI interns that never sleep | Scott Wu (Cognition CEO)
Summary
Scott Wu, CEO of Cognition and creator of Devin, explains how his 15-engineer team uses Devin to merge several hundred PRs per month and demonstrates it live on the podcast. His central thesis: AI shifts engineers from "bricklayers" to "architects" — the valuable skill is precisely defining problems and specifying architecture, while 90% of rote implementation becomes delegable.
Key Takeaways
Give AI agents "tasks, not problems": Devin performs best on well-scoped, verifiable work — for open-ended problems, break them down first and steer at the scoping stage.
Work with multiple agents asynchronously: most engineers at Cognition run 4-5 Devin instances in parallel, checking in at key decision points.
Learn to code anyway: understanding abstractions matters even if you're not writing boilerplate — it lets you peel back layers when agents go wrong.
Jevons Paradox applies: as code generation gets cheaper, total demand for software will expand faster than efficiency gains shrink headcount — expect more programmers, not fewer.
Use agents for onboarding: auto-generated codebase wikis are valuable for new engineers who want to understand a codebase without feeling embarrassed to ask.
Notable Quotes
“It is funny with AI and especially because I would say one of, I would say the most common pieces of advice out there I would say is focus on a really niche cohort. Do things that don't scale, make one use case that's really great and then you grow from there. And I think that's great advice across the board. But yeah, it's kind of interesting because I think with generative AI, you naturally see this where a lot of product experiences can turn out to be more general. And so it's an interesting trade-off for us. This is something that we still always go back and forth on and how much do we want to do more to support all the other kind of use cases out there to handle other things that folks might want to do with Devin.”
“What is a database and how should you think about a database? What is a garbage collection system and how do those work and all of these different pieces? The reason I think that's important is because it's the same with a lot of these other... Arguably we've already gone through these phases in programming and I think this next one is going to be somewhat faster and somewhat bigger, but in many ways a similar flavor, which is when you work with Python today, obviously, a lot of things are already abstracted away from you.”
“It's funny actually because something that in terms of the wording that we thought a lot about as well is just, we've used the term manager of Devins in the past, which of course I think is a big part of it. But I think that the only thing I would point out here is I think that the bricklayer versus architect is closer to the experience than being a manager. Because I think a lot of the difficulty of management or the reason that people shy away from it is more because of a lot of the various.”