Weco is the platform for autoresearch — the new way to optimize code by letting AI agents autonomously edit, test, and iterate against any measurable metric.

Run experiments on our infrastructure, or connect your own. We provide full observability into every experiment, code change, and metric evaluation — so results are reproducible, not just claimed.

Our open-source agent AIDE powers many of the top autoresearch results, achieving ~4x the medal rate of the next best agent across 75 Kaggle competitions (validated by OpenAI on MLE-Bench).

Weco AI

Org chart

Teams

Offices