The former DeepMind researcher turned winning poker-turned-reinforcement-learning agent into stock trader whose recent Series A pushed his valuation to $500 million.
Three former DeepMind researchers who developed an AI that beat humans at poker are now applying the same technology to stock trading. And their bet looks promising. Prague-based laboratory EquiLibre Technologies is valued at $500 million after closing a Series A round led by Creandum. The size of the investment was not disclosed, but Cameron Sellers, vice president of the company, said it was the company’s largest single investment in its history.
The relationship between poker and finance is simple. Both areas lend themselves to reinforcement learning, a way for agents to acquire skills through rewards. According to Martin Schmid, CEO of EquiLibre, the main advantage of trading and markets is that valuation is very simple. In other words, how much money did the agent make?
The main advantage of trading and markets is that they are very easy to evaluate. In other words, how much money did the agent make?
– Martin Schmidt
These technologies are already being put into practical use, not just in games. In partnership with Tower Research Capital, EquiLibre’s algorithms drive billions of dollars in daily trading volume on the S&P 500 and NASDAQ. According to the company, Agent has achieved stable performance since the launch of the virtual currency market in 2025, and currently has a perfect record on the stock exchange with no losses for several months since its establishment.
Perfect record of zero losses for several months since the start
– Equilibre
Applying AI to quantitative hedge funds puts the firm in an environment where automation is the norm and success can quickly lead to cash. This makes EquiLibre attractive to Creandum. “The total potential market for trading in financial markets is one of the largest on the planet, with hundreds of funds that have generated huge profits over the years, dwarfing the success of most ventures.”
The total market that can be traded in financial markets is one of the largest on the planet, with hundreds of funds that have generated huge profits over the years, making the success of most ventures seem small.
– Cameron Sellers
But Schmidt emphasized that Equilibre positions itself not as a financial company but as “a laboratory first and foremost.” Together with co-founders CTO Rudolf Kadlec and CSO Matej Moravcik, the team focuses on research and creative approaches. “I’m not doing this because I want to make the market more efficient. I’m doing this because we all have a deep passion for building something new that hasn’t existed before. It’s incredibly exciting.”
I’m not doing this because I want to make the market more efficient. I’m doing this because we all have a deep passion for building new things that never existed before, and that’s what makes it incredibly exciting.
– Martin Schmidt
After a year of active development, the three founders decided to return to the Czech Republic – Prague – to build a team within the local IT community and reduce talent attrition compared to Western technology platforms. Schmidt said the move made sense, saying, “It’s easier to keep the talent here than it is to keep chasing new things in San Francisco.”
In the same building as EquiLibre is BottleCap AI, highlighting the growing wave of talented AI startups in the region. In the future, the company plans to expand its computing infrastructure and launch one of the largest computing clusters in Central and Eastern Europe.
In terms of funding, EquiLibre has previously raised a pre-seed round backed by Credo and a seed round led by Blossom Capital, valued at approximately $140 million. Investors including Creandum are now backing a new Series A, highlighting the growing interest in frontier AI projects in the financial sector.
Sellers says that while a post-Series A valuation of $500 million is a significant jump, the RL space in trading is growing rapidly: “When we started a few years ago, a lot of people were skeptical, but now RL is the norm and we believe we have an advantage because of our early start.”
This is not a market where a single player wins.
– Martin Schmidt
Despite the competition and risks, EquiLibre’s journey shows strong investor interest in applying frontier AI to finance. The team hopes that reinforcement learning and optimization strategies will establish a new standard in automated trading, maintaining its role as a laboratory that opens new horizons in technology rather than simply pursuing financial gain.
The company continues to develop its technology and expand its computing capabilities, aiming to strengthen its position in both the region and global investment markets through artificial intelligence and machine learning.
