Today we're excited to introduce Devin, the first AI software engineer.
Devin is the new state-of-the-art on the SWE-Bench coding benchmark, has successfully passed practical engineering interviews from leading AI companies, and has even completed real jobs on Upwork.
Devin is an autonomous agent that solves engineering tasks through the use of its own shell, code editor, and web browser.
When evaluated on the SWE-Bench benchmark, which asks an AI to resolve GitHub issues found in real-world open-source projects, Devin correctly resolves 13.86% of the issues unassisted, far exceeding the previous state-of-the-art model performance of 1.96% unassisted and 4.80% assisted.
Check out what Devin can do in the thread below.
I'm still set to build 50 of these, and i've gotten 6 fully done. Bit by bit, things will get done, even with all the little challenges that get in the way.
There's been so much support, though. I'm so glad i'm not doing this in a vacuum. It's just been a constant war of getting worked up and calming down over and over again. If you are out there commercializing your own projects, my heart goes out to you. You can do it.
Since the 30th of December i have given myself a job 7 days a week of constructing the drum machines up there. It has been a journey. I'm close to shipping the first of these units, and maybe then i can settle down from this stress.
That was a bit of a production that happened off-camera.
This is a clone of the EDP Gnat, named the "Gnot" by it's creators at Michigan Synth Works.
It didn't go together exactly as planned, but at least that wasn't because of anything they did. I'll try and recall:
But what isn't shown is the hours of staring and measuring and guessing and staring at schematics that always comes with these projects when they go wrong. That part and my guesses i left out this time since it all wasn't 'in vitro'.