Patronus AI raises $50M Series B to build simulated environments for AI agent testing
Tags AI · Enterprise · Developer Tools
Patronus AI, founded by former Meta AI researchers Anand Kannappan and Rebecca Qian, raised $50 million in a Series B led by Greenfield Partners with participation from Notable Capital, Lightspeed, Datadog, and Samsung. Total funding reaches $70 million. The startup creates 'digital world models' — replicas of websites and internal systems — where AI agents are stress-tested using reinforcement learning. Revenue has grown 15-fold over the past year. The company is currently focused on software engineering and finance domains where agent performance is verifiable.
Technical significance
As AI agents move from demos to production deployments, the ability to test them across millions of simulated scenarios becomes critical infrastructure. Patronus's 15-fold revenue growth and customer roster spanning virtually every frontier AI lab suggests this is becoming a must-have tool, not a nice-to-have. The approach mirrors how Waymo used synthetic environments for autonomous vehicle validation — now applied to software agents.