From Poker Table to Wall Street: DeepMind Alumni's AI Now Trading Billions
Three former DeepMind researchers who once taught artificial intelligence to master the complexities of no-limit Texas hold'em poker have now applied the same technological principles to a far more lucrative domain: quantitative hedge fund trading . Their Prague-based AI laboratory, EquiLibre Technologies, has reached a valuation of $500 million following a Series A funding round led by Creandum, marking one of the most notable transitions of reinforcement learning expertise from gaming to finance .
The connection between poker and stock trading is not as unexpected as it might initially appear. Both domains involve incomplete information, probabilistic reasoning, and strategic decision-making under conditions of uncertainty . The common denominator is reinforcement learning, an AI training technique where self-learning models are incentivized through a system of rewards and penalties, essentially learning optimal strategies through trial and error . According to Martin Schmid, chief executive officer and co-founder of EquiLibre, financial markets offer an exceptionally clear feedback signal for this approach. The scoring is refreshingly simple, as he puts it, measuring precisely how much money the AI agent has made .
This is not merely a theoretical exercise or an experimental project. EquiLibre has established a partnership with quantitative trading firm Tower Research Capital to deploy its algorithms across major United States stock indices, including the S&P 500 and the NASDAQ . The startup claims its AI agents have been trading billions of dollars in daily volume and have posted what it describes as a perfect record of zero negative months since inception, meaning each calendar month has ended with net positive returns . The system initially launched on cryptocurrency markets in 2025 before expanding to traditional stock exchanges .
The founding trio of Martin Schmid, Rudolf Kadlec, and Matej Moravcik first came together as visiting PhD students at DeepMind's Edmonton office in Alberta, Canada, which was Alphabet's first international artificial intelligence research location . During their time there, they built DeepStack, the first AI program to defeat professional players at no-limit Texas hold'em . They also collaborated with professors who now serve on EquiLibre's advisory board, including Rich Sutton, who received the Turing Award in 2024 for his foundational contributions to reinforcement learning .
Rather than remaining in North America or relocating to traditional technology hubs, the founders chose to return to their home country, the Czech Republic, establishing their operations in Prague . This decision was driven by the availability of talented colleagues and friends from the Czech diaspora who had worked at Google and other prominent technology companies . The startup built its initial team of twenty-five people starting in 2022, and Schmid believes the location continues to provide advantages, noting that the absence of constant hype cycles makes it easier to retain talented personnel compared to locations like San Francisco .
EquiLibre explicitly positions itself as a research laboratory first rather than a financial firm . Schmid emphasized that he and his co-founders are not motivated by a desire to optimize market efficiency but rather by the excitement of building novel systems that have never existed before . This lab-first identity resonated with investors, with Cameron Sellers from Creandum noting that EquiLibre's founders bring a rare combination of deep reinforcement learning expertise and a willingness to apply it to a domain where automation is already well established . The venture capital firm described the investment as the largest single commitment it has ever made in one go into a company .
The startup plans to use its Series A funding to scale its compute infrastructure, aiming to build what it expects will be one of the largest computing clusters in Central and Eastern Europe . According to available data, EquiLibre previously raised a $10 million seed round led by Blossom Capital at a $140 million valuation, with pre-seed backing from Credo . Reinforcement learning has gained broader acceptance in trading since EquiLibre was founded, with Schmid acknowledging that initial skepticism has given way to recognition of the approach's potential . However, competition is intense, with major trading firms like Jane Street already using reinforcement learning alongside large language models and claiming access to tens of thousands of high-end graphics processing units . EquiLibre aims to differentiate itself by extracting more performance from fewer chips, pursuing efficiency rather than simply scaling hardware .
Schmid does not view the market as a winner-takes-all competition, noting that there is room for multiple successful players in this space . The startup's goal is to be recognized as the leading AI laboratory in trading, building a reputation based on research output and consistent returns rather than attempting to capture every dollar in the market . This pragmatic perspective reflects the founders' academic and research-oriented background, suggesting that their approach to financial markets is driven more by intellectual curiosity and technical challenge than by purely commercial considerations.
