Docs / Company
Retired
What is Patronus AI
The LLM evaluation platform documented on this site is retired and no longer offered. Patronus AI is a behavioral research lab for AI. We study where models fail and how to improve them.
Patronus AI is a research lab developing training data and simulated environments for frontier AI. Its work spans reinforcement learning environments and digital world models for agents operating in software.
Mission
Research and engineering under one roof
Patronus AI is a behavioral research lab for AI. We build worlds for models to explore and study what they do there and why. Our work combines open-ended research with hands-on engineering to meet difficult problems on the frontier of AI.
- Founded
- 2023 by Anand Kannappan and Rebecca Qian
- Headquarters
- San Francisco
- Team
- Researchers and engineers from Meta Superintelligence Labs, Amazon AGI, Apple and Google
- Series B
- $50M, June 2026, led by Greenfield Partners
Domains
Every domain asks for different judgment
- AlignmentFollow instructions and know when to push back.Domain → (opens in a new tab)
- CodingShip a feature in code no one documented.Domain → (opens in a new tab)
- Knowledge WorkMake one decision from sources that disagree.Domain → (opens in a new tab)
- ScienceIdentify and explore the untested question.Domain → (opens in a new tab)
- Computer UseGet work done in software built for people.Domain → (opens in a new tab)