1 repository on SrcLog
Add it to a GitHub README or profile β the card updates as the ranking changes:
Optimize LLM agent harnesses through tunable Formulas to dynamically enhance the agent, and improve those Formulas with optimizers based on collected agent rollout trajectories.