AI automation companies, somewhat predictably, seem to be using AI to write their own code.
In a recent interview, Scott Wu, CEO of Cognition, the company behind the AI coding tool Devin, offered a fascinating glimpse into the future of software development. His comments reveal not only the significant impact Devin is already having on Cognition’s internal workflow but also paint a picture of a drastically different landscape for programmers in the coming years. One particularly striking revelation is the sheer volume of pull requests already being generated by Devin.
“Okay, so today about a quarter of our PRs are Devin’s. Where do you think this will be at the end of the year? What would you guess?” the interviewer asked.
Wu responded, “I think by the end of this year, we expect it to be more than half. And I mean, as time goes on, you know, one of the things that we’ve seen is you’re able to do more and more work asynchronously, right? And you’re able to hand off more and more. I think the sole [focus] that has really been, basically, just about defining the problem that you’re facing and really thinking through exactly what is the solution that you want to build. You know, thinking through the architecture or thinking through the details and really kind of mapping out in your mind exactly what you want to build, basically, and what you want to have your computer do.”
He continued, “And one of the ways that we’ve kind of thought about Devin and building Devin is really allowing engineers to go from bricklayer to architect, so to speak. And a lot of it is getting to the point where you can do the high-level directing and you can basically specify things exactly how you want. I think it’s very much about still having the human in control and having the human able to do the full specification, but just multiplying the magnitude of what you can do and what you can build in one day or one hour or however long. And you know, someone from 50 years ago might already call Python… you know, you just get to explain in English what you want and now the computer does it for you, right? And that’s great, and I think it’s really powerful.”
“I think there is going to be way more product and maybe way more programmers and way more engineers a few years from now than there are today. And I think pretty quickly the form factor of what it means to be a programmer is going to change. And in some sense, it already has, but I think there’s just going to be so much more for us to build,” he said.
“Going back to the way you guys use Devin, so you said that every engineer has kind of this fleet of Devins. How many Devins per engineer do you find most people are working with these days at your company?” the interviewer asked.
“Yeah, so it’s very asynchronous and so obviously you can kick them up and start them up and shut them down basically as you see fit. But, but most folks on the team are often working with up to five Devins at once, I would say. “
“And then how many engineers do you guys have?”
“Our engineering team today is about 15 people.”
“So there’s five times the number of Devins as engineers. What I love about this is this is just like a glimpse into where the future is going. You guys are so ahead of how companies work with AI engineers and so seeing how you operate is going to be…it’s essentially how most companies will end up operating,” the interviewer said.
Devin operates differently from AI coding IDEs like Cursor and Windsurf. It creates a fully autonomous coder that reads messages on Slack and independently generates and commits code. Devin is also significantly more expensive than Cursor or Windsurf, and costs $500 per month.
And this ties in with Wu’s vision of turning engineers into architects, with AI systems like Devin doing the actual coding. And the company seems to be using its own product — with a quarter of their pull requests already originating from Devin and projections suggesting this will exceed half by year’s end, systems like Devin could fundamentally change how programming is done. The shift from “bricklaying” to architectural design suggests a dramatic increase in developer productivity. This could lead to an explosion in software development, potentially fueling innovation across industries. And this dynamic, where relatively small teams of human engineers oversee armies of AI agents, could reshape the entire software development landscape sooner than we think.