Three former DeepMind researchers who once built an AI that defeated human poker professionals have now applied the same technology to stock trading, and the bet appears to be paying off. Their Prague-based AI lab, EquiLibre Technologies, is now valued at over $500 million after closing a Series A funding round, as reported by TechCrunch. The round was led by Creandum, which described it as the largest single investment the firm has ever made in one go into a company.
Reinforcement learning bridges poker and Wall Street
The common thread between poker and financial markets is reinforcement learning, a training technique where self-learning models are incentivized by rewards. EquiLibre CEO Martin Schmid notes that the scoring in trading is straightforward: how much money did the agent make? This is not just play money. In partnership with quant firm Tower Research Capital, EquiLibre's algorithms have been trading billions in daily volume across the S&P 500 and Nasdaq. The startup claims its agents have maintained a perfect record of zero negative months since inception, meaning each month has ended with overall positive returns, starting with crypto markets in 2025 and now expanding to stock exchanges.
Sponsored Protocol
A $500 million valuation and a team far from Silicon Valley
Creandum invested because the total addressable market of trading in financial markets is enormous, and many funds have generated quantum profits. However, Schmid emphasizes that EquiLibre defines itself as a lab first, not a finance firm. The founders — Schmid, CTO Rudolf Kadlec, and CSO Matej Moravcik — lack a finance background; their drive is the excitement of building new things. They chose to return to their home country, Czechia, leveraging the Czech diaspora at Google and elsewhere. The current team of 25 is easier to retain in Prague than in San Francisco, where new AI ventures pop up every two months. EquiLibre plans to scale its compute infrastructure, aiming to bring online one of the largest compute clusters in Central and Eastern Europe.
Sponsored Protocol
The rise of reinforcement learning and competitive challenges
When EquiLibre started in 2021, reinforcement learning faced skepticism; now it is the standard. The startup must watch out for giants like Jane Street, which already uses RL combined with large language models and owns tens of thousands of high-end GPUs. EquiLibre aims to achieve more with fewer chips. Schmid concludes that this is not a winner-takes-all market, but playing the right cards will be crucial. For more on AI advancements in specialized fields, check our article on Anthropic Unveils Claude Sonnet 5. To understand reinforcement learning fundamentals, see the Wikipedia page on reinforcement learning.
Sponsored Protocol
