creators

Noam Brown: OpenAI researcher on reasoning and self-play

Noam Brown is a research scientist at OpenAI working on reasoning and self-play, and built the Libratus, Pluribus and CICERO game-playing AI systems.

Noam Brown is a research scientist at OpenAI, where he works on reasoning, reinforcement learning, self-play and multi-agent AI; his own CV dates the post from 2023 to the present, and his site records that he was "a foundational contributor to the development of reasoning models (e.g., o1 / strawberry)". He took a BA in mathematics and computer science at Rutgers, then a master's in robotics and a PhD in computer science at Carnegie Mellon, where with his advisor Tuomas Sandholm he built Libratus and Pluribus, poker programs that beat top human professionals at heads-up and six-player no-limit Texas hold'em. Libratus received the Marvin Minsky Medal for Outstanding Achievements in AI; Pluribus was the cover article of Science in 2019 and a runner-up for the magazine's Breakthrough of the Year. Brown was named to MIT Technology Review's 35 Innovators Under 35 in 2019. Before his research career he worked as an algorithmic trading engineer and then as a research assistant at the Federal Reserve Board. From 2018 to 2023 he was a research scientist at Facebook AI Research in New York, where he co-created CICERO, the first AI agent to reach human-level performance in the natural-language strategy game Diplomacy. He gave an invited talk at ICLR 2026 on the implications of large-scale test-time compute.

Specialization

reasoning, reinforcement learning, self-play, multi-agent AI

Country

United States