Coordination for Scale RL (casRL)
2020–PresentSole creator and principal architect
An open-source research framework that builds RL and MARL experiments from reusable implementations of policies, rollout collection, replay, objectives, optimization, and model updates. casRL implements more than 100 algorithms and variants across online, offline, hybrid, single-agent, and multi-agent learning and sustains 2–4 million environment steps per second during end-to-end single-node training on one CPU and one GPU.