AI Evaluation Engineer – Reinforcement Learning & Agents

Job Description

• Build evaluation environments, post-training systems, and infrastructure for long-horizon enterprise AI agents.
• Define tasks, states, tools, graders, and reward signals; develop rollouts, orchestration, trajectory inspection, and grader pipelines.
• Measure agent correctness and efficiency, investigate evaluation failures and reward-quality issues, and build production-quality systems.
• Requires hands-on experience with AI environments, evaluations, reinforcement learning infrastructure, or related agent-training systems, plus strong software engineering and infrastructure delivery skills.
• Full-time, primarily on-site role anchored in San Francisco, CA; open to exceptional new graduates, early-career engineers, and senior candidates. A PhD is not required.