Kapso
Open-source engine from Leeroo that runs experiments toward a goal you define, deploys the best result and keeps graded lessons from each run.
- AI agent framework
- Free
Pricing
Free and open source (MIT). Optional Leeroopedia knowledge service comes with $20 of free credit on a new account; model usage is billed by your provider.
Overview
Kapso is an open-source Python engine from Leeroo, installed as leeroo-kapso. You give it a goal and a way to measure success, and it tries a series of approaches, scores each attempt and deploys the best solution on your own stack. After a campaign it reviews what happened and saves lessons as knowledge cards with a confidence score and a note on when they apply, kept in a local git repo on your machine unless you share them. It is not tied to one model or coding agent, and it can look things up in Leeroo's Leeroopedia knowledge base for ML and AI frameworks. You get a tested, deployed solution plus a growing bank of lessons for the next run.
Key features
- Runs scored experiments toward a goal
- Deploys the best attempt
- Saves graded lessons as knowledge cards
- Lessons stay in a local git repo
- Not tied to one model or agent
- Optional Leeroopedia knowledge lookups
- Open source under MIT
Use cases
- Experiment tracking
- AI experimentation
- Self-improving agents
- Machine learning engineering
- Model optimization
Show all 7Show fewer
Alternatives
Last updated 2026-11-13