ESTABLISHED AI Agents

Reef

Continual learning infra for self-improving agents

Why it matters now: Continual learning infra for self-improving agents

Editor's take

Reef is an open-source infrastructure for continual self-improving agents. It connects agent inference, feedback, learning, and versioned delivery. Reef can be used to train model weights or improve an agent’s harness, including its prompts, rules, and skills. The infrastructure is designed to keep improving simply by learning from how you interact with your agent.

Who is it for

  • Those who want their agent to keep improving simply by learning from how you interact with your agent

Requirements

A trainable model, a supported GPU stack, and feedback your recipe can use