rhesis

rhesis is an open-source testing platform designed for AI teams. It brings together engineers, product managers, and domain experts to generate test cases, simulate adversarial conversations to uncover weaknesses, and trace every failure back to its root cause. This collaborative approach aims to improve the robustness and reliability of AI applications.

Key advantages include cross-functional collaboration, which allows non-technical stakeholders to contribute to test design, and adversarial simulation that enhances safety and boundary testing. As an open-source project, it can be freely inspected, modified, and integrated into existing workflows.

However, the publicly available evidence is extremely limited—only a basic GitHub repository description with no stars, forks, or documentation. The project's maturity, real-world viability, and stability are unverified. It may lack features, maintenance, or contain unknown risks. Deployment requires technical expertise, and no managed service exists.

This tool may suit AI teams with in-house engineering capabilities who are willing to experiment and customize their testing stack.