Patronus AI, a San Francisco-based startup that builds simulated environments to evaluate AI agent performance, announced a $50 million Series B funding round on June 25, 2026. The round was led by Greenfield Partners, with participation from Notable Capital, Lightspeed, Datadog, and Samsung, bringing the company’s total funding to $70 million.
Founded in 2023 by former Meta AI researchers Anand Kannappan and Rebecca Qian, Patronus creates what it calls “digital world models” — replicas of websites and internal systems where AI agents are stress-tested using reinforcement learning. The approach iteratively rewards agents for completing tasks correctly and penalizes errors, helping model makers identify and close performance gaps before deployment.
The company’s revenue has grown 15-fold over the past year. Glenn Solomon, a managing director at Notable Capital, said virtually every frontier AI lab and many emerging startups are now customers, describing demand as “nearly insatiable.”
The core problem Patronus addresses is the gap between benchmark scores and real-world reliability. High scores on agent-oriented benchmarks do not guarantee that an AI can correctly complete complex tasks such as booking travel or conducting financial analysis. Patronus compares its simulation approach to how Waymo tested autonomous vehicles in synthetic environments against rare hazards before road deployment. Solomon noted that AI agents tend to take shortcuts, and that Patronus is “really good at spotting the hacks and making sure they are holding the models accountable.”
The company currently offers its simulated environments for software engineering and finance use cases. Kannappan indicated those are just a starting point, with ambitions to support agents running continuously for days or weeks and to expand into areas that are harder to verify. Patronus says its primary competition comes from the internal evaluation teams that AI labs have already built, rather than from other external vendors.
Source: TechCrunch