Senior Research Scientist, Agentic AI

Own and develop the evaluation platform pipelines that run AI agent traces through model-based judges at scale.

Design and implement evaluation methodologies and benchmarks for agent reasoning, planning, tool use, reliability, and safety.

Take research problems from concept to prototype to shipped feature, ensuring evaluation results are actionable for teams.

5+ years experience in machine learning, applied AI, prompt engineering, or agentic AI with real user-facing product delivery.

Strong proficiency in Python programming.

Practical experience with agentic AI concepts: planning, reasoning, memory, tool use, retrieval, and long-context handling.

Experience designing evaluation methodologies, not just conducting evaluations.

Demonstrated end-to-end ownership of AI evaluation systems and pipelines, including scoring logic and data curation.

Hands-on expertise with LLM APIs and prompt engineering for production environments, addressing cost and latency tradeoffs.

Experience communicating technical concepts effectively to both technical and non-technical stakeholders.