Eve Security raises $4.5M seed extension for AI agent governance

Developer infrastructure that turns unstructured inputs into typed decisions and probability scores for automated software workflows.

Provides crowdsourced AI model performance leaderboards

Reinforcement-learning environments that help AI labs train models to solve tasks without exploiting flaws in the grading system.

Open-source AI models and the Hermes agent, which retains project memory and reusable skills across tasks.