2025
Cognition published a post about how much its Devin product improved with Sonnet 4.5.
“They put out a post yesterday about how much their their product devin improved with the model and some of the work they did to to make that product better.”
Topic
Devin
10 facts from 3 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.
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
Cognition acquired Windsurf, so its products now include Devin, an AI coding agent, and Windsurf, an IDE.
“most recently just acquired Windsurf. And so, you know, Devin, the agent, and Windsurf, the IDE.”
Undated
Scott Wu describes Devin as an asynchronous product in which a user delegates a task to an agent, unlike IDE tools that speed up typing.
“Devin is a very different paradigm of what I would call like an async experience, right? Where you have an agent and you delegate a task.”
Undated
Scott Wu says Devin's customers range from large banks such as Goldman Sachs and Citibank to startups with two or three people.
“We work with some of the biggest banks in the world, like Goldman and Citibank, all the way down to startups with two or three people.”
Undated
Scott Wu says that in successful organizations Devin typically merges 30 to 40 percent of all pull requests.
“Typically, in a successful org, Devin is merging something in the range of 30 % to 40 % of all the pull requests that come through.”
Undated
Scott Wu says customers who measure enterprise code migrations internally report an 8 to 15 times productivity gain with Devin.
“when folks measure internally, they see something like an 8 to 15x gain for a lot of these use cases with Devin”
Undated
Scott Wu says the Cognition team uses Devin to order DoorDash and Amazon packages.
“we order our DoorDash on Devin. You know, we order our Amazon packages with Devin.”
Undated
Scott Wu says a Cognition colleague named Walden used Devin to negotiate an airline refund for a canceled flight, and it succeeded.
“Walden had a flight that got canceled and was trying to, you know, use Devin to go and, like, negotiate with the airline to get the refund for it.”
Undated
Cognition's engineering interview has candidates build their own version of Devin, a full agent, in about eight hours.
“basically just having people build their own Devin in eight hours and seeing how far they get with it.”
Episodes