Kiln-AI releases Harness Tuner for agent harness tuning
Start here
Kiln-AI released Harness Tuner, an open-source CLI that tunes an agent's tools, prompts, skills, model and settings against existing evals. On one production multi-turn agent with over 200 tools, the team reports 2.5x fewer failures, 6x lower cost and 2x lower latency on held-out data using a smaller model. The tool runs on Claude Code, Codex or OpenCode subscriptions and keeps train, val and test splits enforced in code. It works with the free Kiln app for experiment tracking. Harness Tuner itself is free and open source but requires LLM API calls: users add an OpenRouter API key to Kiln, and a tuning run costs about $2.
Generated from reports · updated 11 min ago
Report timeline
Follow the reports to see the event from different sides.
- Hacker News · AIKiln-AI releases Harness Tuner, an eval-driven tuning loop for agent tools, prompts and models
Kiln-AI released Harness Tuner, an open-source CLI that tunes an agent's tools, prompts, skills, model and settings against the evals you already have. On one production multi-turn agent with over 200 tools, the team reports 2.5x fewer failures, 6x lower cost and 2x lower latency on held-out data with a smaller model. It runs on Claude Code, Codex or OpenCode subscriptions, keeps train, val and test splits enforced in code, and works with the free Kiln app for experiment tracking.
Heat for this event
Not enough continuous observations yet to draw a trend.