As AI agents become more capable, businesses face a new challenge. They need to know whether those systems can perform reliably before handling real work.
Patronus AI wants to solve that problem. The startup has raised $50 million in Series B funding to expand its AI evaluation platform.
The round was led by Greenfield Partners, with participation from Notable Capital, Lightspeed, Datadog, and Samsung. The investment brings the company’s total funding to $70 million.
Testing AI Before It Goes Live
AI agents are moving beyond simple chatbots. They can now write code, analyze financial data, and complete multi-step business tasks.
That makes testing more important than ever. Companies want confidence that AI agents will perform correctly in real-world situations.
Patronus AI builds simulated digital environments for that purpose. These environments allow AI agents to complete tasks before they are deployed inside an organization.
The platform measures how agents respond to different scenarios. It also identifies errors that may not appear in traditional benchmark tests.
Founded by Former Meta AI Researchers
Patronus AI was founded in 2023 by former Meta AI researchers Anand Kannappan and Rebecca Qian.
The founders believe benchmark scores alone cannot predict how AI agents will perform in real business workflows.
Instead, they created a platform that evaluates AI behavior inside realistic digital environments. This approach gives developers deeper insight into an agent’s reliability.
Digital Worlds for Enterprise AI
The company calls these environments “digital world models.”
They recreate websites and enterprise systems where AI agents can safely perform tasks. Developers can then observe how the agents respond under different conditions.
The goal is to find problems before deployment. That helps organizations reduce risk as AI takes on more responsibility.
Patronus AI says its platform can also detect when AI agents take shortcuts instead of completing tasks correctly.
Strong Demand Across the AI Industry
The company says many leading AI labs and emerging startups already use its technology.
According to Glenn Solomon, Managing Director at Notable Capital, demand for these testing environments continues to grow.
That momentum has fueled rapid business growth. Patronus AI says its revenue increased fifteen-fold over the past year.
Looking Beyond Software and Finance
Today, the company mainly supports software engineering and financial use cases.
However, CEO Anand Kannappan says the long-term opportunity is much larger. Future digital environments could support many other enterprise workflows.
Rather than relying on human reviewers, Patronus AI evaluates how agents behave automatically inside simulated environments.
Building Trust in AI Agents
As enterprises deploy more autonomous AI systems, evaluation is becoming a core part of AI development.
Patronus AI is positioning itself as the infrastructure behind that process. Its platform helps companies identify issues before AI agents interact with customers or critical business systems.
With fresh funding and growing adoption, the company is expanding its role in helping enterprises deploy AI agents with greater confidence.