Analysis
EquiLibre Technologies, a Prague-based AI lab founded by three former DeepMind researchers, raised a Series A led by Creandum at a $500 million valuation, according to TechCrunch. Creandum vice president Cameron Sellers confirmed the round was the largest single investment the firm has ever made into one company, an unusually strong signal from a well-established European venture firm.
The founding team -- CEO Martin Schmid, CTO Rudolf Kadlec and CSO Matej Moravcik -- met as visiting PhD students at DeepMind's first international research office in Edmonton, Alberta, where they built DeepStack, the first AI program to defeat professional players at no-limit poker. That result was a landmark in reinforcement learning because poker, unlike chess or Go, involves hidden information and bluffing, requiring the AI to reason under genuine uncertainty rather than perfect information.
The founders' insight was that poker and financial trading share a core structural similarity: both are well suited to reinforcement learning, where self-learning models improve through reward-based feedback rather than labeled training data. EquiLibre applies its reinforcement-learning algorithms to trade stocks and cryptocurrencies in partnership with quantitative hedge funds, including Tower Research Capital, one of the more established and technically rigorous firms in systematic trading.
The competitive landscape in AI-driven quantitative trading includes established players like Renaissance Technologies, Two Sigma and Citadel building proprietary models in-house, alongside a newer wave of AI-native trading startups. EquiLibre's differentiation is deep reinforcement-learning expertise proven in an adversarial, imperfect-information domain -- a genuinely rare research pedigree that most trading-focused startups don't have.
For founders and researchers, EquiLibre is a template for how narrow, prestigious research achievements (a specific AI breakthrough in a specific domain) can translate into venture-scale businesses when the underlying technique generalizes to a lucrative adjacent market. For LPs evaluating deep-tech-to-fintech crossover bets, a hedge fund of Tower Research's caliber as a paying customer is a much stronger signal than typical early trading-AI pilot deals.
The bear case is that trading strategies which work in backtests or early deployment can degrade quickly as markets adapt, and quant trading is notoriously difficult to sustain at scale even with strong technical pedigree -- alpha decays, and competitors copy successful approaches fast. What to watch: EquiLibre's assets under management or trading volume with Tower and other partners, whether it expands beyond partnership models into its own fund, and how its performance holds up across different market regimes.
