google-deepmind/envlogger

Trace mining for post-training. Filter trajectories, draft verifiers, export SFT and RL data.

OrcaReplay — Time travel for AI agents. Record, replay, fork, and debug any agent run with any model. Built by the OrcaRouter.ai team.

Convert sessions across harnesses to a unified trajectory format - designed to be consumed by agents (e.g. for memory formation, dreaming, search)

A lightweight, local-first, and free experiment tracking library from Hugging Face 🤗

Record and Replay AI Agent Trajectories.

A plug-in debugger and visualizer for RL reward functions. Detects reward hacking, tracks training health, and renders a live terminal dashboard.

Reverb is an efficient and easy-to-use data storage and transport system designed for machine learning research

Build RL environments for any domain from real cases that Browser Use agents mine from the web

Observability for AI agents. See what your agent did, why it cost that much, and what to fix.