Reinforcement learning environments built from financial market data for training AI agents toward recursive self-improvement.
EdotEnv builds scalable, increasingly difficult RL environments using financial market data to enable AI agents to continuously improve through recursive self-improvement. The environments are multi-step, requiring agents to form hypotheses, design experiments, verify results, and iterate on progressively harder tasks. The company works with frontier AI labs and academic groups building research harnesses, evaluation benchmarks, and post-training environments. Markets create natural competitive pressure that forces genuine capability improvement. EdotEnv is a product of EdotEnv.