DeepMind alumni apply poker AI techniques to quant trading

TechCrunch reported that Prague-based EquiLibre Technologies, founded by three former DeepMind researchers behind a poker-playing AI, has raised a Series A at a $500 million valuation. The company applies reinforcement learning to trading and says its algorithms, used with Tower Research Capital, trade billions in daily volume across the S&P 500 and Nasdaq with no negative months since inception. The round was led by Creandum, which called it the largest single investment the firm has made into one company, while EquiLibre plans to scale compute infrastructure in Central and Eastern Europe.

Key details: EquiLibre was founded by former DeepMind researchers tied to DeepStack poker AI, The startup is valued at $500 million after a Series A led by Creandum, Its agents are trading billions in daily volume with Tower Research Capital.

Why it matters: The story shows reinforcement learning moving from game benchmarks into high-stakes markets, where performance claims, compute efficiency, and risk controls will be tested with real capital.