Head of Evals

Lead AI safety evaluations for frontier models, ensuring red-teaming data quality and building evals that measure model safety.
- Review red-teaming tasks and transcripts to maintain quality standards; automate judgment patterns into agent skills and pipeline automation.
- Design new evaluations to test model robustness, identify failure modes, and communicate safety findings to frontier labs.
- Set evaluation strategy and prioritize which safety tests Trajectory should build next to scale data collection and model assessment.
- Requires 2+ years software engineering experience and demonstrated ability to use LLMs and coding agents to automate workflows independently.
- Desirable: experience building evals, benchmarks, grading pipelines, or familiarity with prompt injection, red teaming, or AI safety research.

Trajectory Labs is a startup that builds RL environments for frontier AI labs to train robust, secure, and reliable models.