Ex-DeepMind Trio's EquiLibre Hits $500M Valuation in AI Trading

EquiLibre Technologies, a Prague-based AI lab founded by three former DeepMind researchers, has reached a valuation of $500 million following a successful Series A funding round. The startup, which gained recognition for building an AI capable of defeating human professionals at no-limit poker, has now pivoted its reinforcement learning technology toward the financial markets. The investment round was led by Creandum, marking the largest single investment the firm has ever executed.

From Poker Tables to Wall Street

The founding trio—CEO Martin Schmid, CTO Rudolf Kadlec, and CSO Matej Moravcik—originally developed their expertise while working as visiting PhD students at DeepMind’s research office in Edmonton, Alberta. During that time, they created DeepStack, the first AI program to beat professional players at Texas hold ’em. Their work was supported by high-profile advisors, including Turing Award winner Rich Sutton.
EquiLibre is now applying these same reinforcement learning techniques to stock trading. By partnering with the quant firm Tower Research Capital, the startup’s algorithms are actively trading billions in daily volume across the S&P 500 and Nasdaq. According to the company, their agents have maintained a perfect record of zero negative months since they began operating in crypto markets in 2025 and subsequently moved into stock exchanges.

A Focus on Research Over Finance

Despite their success in the financial sector, the founders maintain that EquiLibre is a lab first, not a finance firm. Schmid noted that the team does not have a background in finance and is primarily driven by the challenge of building novel technology rather than market efficiency. This technical focus is what attracted investors like Creandum, who see the massive potential of applying advanced AI to global financial markets.
The company has grown to a team of 25 people since its inception in 2022. By basing its operations in Prague, EquiLibre has been able to leverage a strong local talent pool and maintain a stable workforce. The startup is now focused on scaling its infrastructure, with plans to bring online one of the largest compute clusters in Central and Eastern Europe.

Navigating a Competitive Landscape

The field of AI-driven trading is increasingly crowded, with major players like Jane Street already utilizing reinforcement learning and large-scale GPU infrastructure. While EquiLibre aims to "get more from less" by optimizing its compute usage, the founders acknowledge the competitive nature of the industry.
However, Schmid remains optimistic about the company's position. Having started their work in reinforcement learning four years ago, the team believes they hold a strategic advantage. As they continue to refine their models, they aim to solidify EquiLibre’s reputation as the premier AI lab in the trading space, operating under the belief that the financial market is not necessarily a winner-takes-all environment.