Andrew Gritsevskiy is Co-Founder of RunRL, a Y Combinator-backed platform delivering reinforcement learning as a service to improve LLMs and AI agents, based in San Francisco, California.

Andrew studied at the University of Toronto and UCLA, interned at MIT, and was a PhD student in reinforcement learning at the University of Wisconsin-Madison before dropping out to co-found RunRL.

Andrew has deep expertise spanning reinforcement learning, AI alignment, quantum information, neuroscience, and computational biology, and has co-authored peer-reviewed research published in Nature Machine Intelligence and Transactions on Machine Learning Research.

Prior to RunRL, Andrew co-founded Cavendish Labs, a Vermont-based research lab addressing existential risks, and served as Principal Researcher at Contramont Research, an AI safety lab focused on LM backdoors, real-world evaluations, and scalable oversight.

RunRL, part of YC Batch X25, has attracted investment from Y Combinator, Macroscopic Ventures, and Pioneer Fund, and serves enterprise clients across domains including chemistry research and document processing, enabling models to self-improve through custom reward functions.