⬢github Python · 215 ★ +76 since we first saw it · pushed 19 h ago · MIT
Leeroo-AI/kapso
Kapso by Leeroo: an engine that runs experiments toward an objective, deploys the best solution on your stack and learns from every run into a bank of lessons with a credence score and a boundary. First on RelBench and ALE-Bench, top open-source on MLE-Bench, above every human contestant at IOAI 2026.
Kapso is a Python engine that autonomously runs experiments toward a stated objective, deploys the best solution on your infrastructure, and turns each run's outcomes into a bank of 'lessons' with credence scores and applicability boundaries that inform future campaigns. It targets benchmarks like RelBench, ALE-Bench, MLE-Bench, and offers a plugin for AI coding agents.
Why now: It was featured on Show HN and claims headline results: first on RelBench leaderboards and scoring above every human contestant at IOAI 2026's autonomous AI track.
Who it is for: ML engineers and researchers who want automated experimentation and model deployment pipelines over their own data and stack.
ai-agentsale-benchautoresearchcontinual-learningdeploymentexperimentationharnessioaillm-agentsmachine-learning
Stars over our 172 snapshots: 139 to 215, since 1 d ago.
Where people talked about it
API: https://socialmediatrends-api.osmike.com/v1/repos/Leeroo-AI/kapso