Launching today
Remy
AI engineer that takes a task from plan to tested code
1 follower
AI engineer that takes a task from plan to tested code
1 follower
Remy is an AI software engineer. Give it a bug, a feature or a migration and it reads your codebase, plans the change, writes the code, runs the tests, then opens the app or browser to confirm it works. If a check fails, it fixes it and tries again. Remy works across leading frontier and open-weight models. Pin one yourself, or let Dynamic pick the right model for each task and escalate when it stalls. Runs on desktop, CLI, web and mobile with the tools you already use. Now in private beta.
Free Options
Launch Team
AWS Startup AdvisorPersonalized AWS guidance that meets you where you’re already building
Promoted
Hi Product Hunt, I’m Erin, PM on Remy at Corepass.
We built Remy because every AI coding tool we used stopped at the same place: it wrote the code, and then the rest was on us. Run the tests, open the app, find out it didn’t actually work, go back. We wanted an engineer that finishes the job.
So Remy takes a task end to end. It reads the codebase, plans, writes the code, runs the tests, and then opens the app or the browser to check the result the way a person would. If something fails, it fixes it and tries again. You get a verified change, with the evidence attached.
Two things we think are different. Remy isn’t tied to one model. It runs across the leading frontier and open-weight models, and our routing layer, Dynamic, picks the model for each task and escalates when a task stalls. And Remy works wherever you do: desktop, CLI, web and mobile, with the same projects everywhere.
We’re in private beta. Comment with your email or send me a message and I’ll get you a code with 30 days of Remy Max.
The feedback that would help us most: give Remy a real task from your own repo, not a toy, and tell us where it got stuck or asked you something it should have figured out on its own. That’s exactly what we’re fixing right now.
Thanks for taking a look.