Launching today
Remoroo
Runs overnight experiments on your code locally
29 followers
Runs overnight experiments on your code locally
29 followers
Remoroo is an autonomous experimentation agent for code. Give it a repo, a metric, and a time budget. It edits code and configs, runs tests, benchmarks, and training jobs, keeps changes that improve results, reverts the rest, and leaves you with verified patches, benchmark deltas, and artifacts in the morning. Built for teams doing long-running engineering and research workflows.
Free Options
Launch Team / Built With
Hey Product Hunt.
I’m Adham, founder of Remoroo.
I built this because too much engineering work still looks the same: tweak code, wait for a long run, get worse results, revert, try again. The hard part often is not writing the patch. It is babysitting the experiment loop and proving whether anything actually improved.
So we built Remoroo to run that loop autonomously on your machine. You give it a repo, a measurable goal, and a time budget. It plans, edits code and configs, runs training jobs, tests, or benchmarks, evaluates against the metric, keeps changes that improve results, and reverts the rest.
The big shift for us while building it was realizing we did not want another coding copilot. We wanted something that could take ownership of long-running experiments and leave behind proof: verified patches, benchmark deltas, and artifacts you can inspect in the morning.
We would especially love feedback from teams working on ML, research, robotics, performance tuning, or any workflow where a single experiment can take minutes or hours.
What would make you trust an agent to run repeated experiments on your codebase?