Agent Researcher

Bunkerhill Health•San Francisco, CA

6h•Onsite

About The Position

Bunkerhill builds long-horizon agents for health system knowledge work. Its agents support tasks performed consistently across customers as well as tasks tailored to individual health systems. The team works in person from offices in San Francisco and New York City and emphasizes building for the patient, taking ownership and shipping, and being bold together. The Agent Researcher helps manage the lifecycle of agents for closed-ended tasks that require repetitive intelligence, from understanding customer pain points through launching an agent that runs across patients. The role focuses on informing agent design through experimentation, improving correctness, cost, and velocity, enabling agent developers, creating reliable third-party connectors, owning testing and evaluation, advocating for platform improvements, and identifying opportunities for repetitive-intelligence automation.

Requirements

Strong technical foundation, including comfort writing code, collaborating in technical teams, and understanding agent concepts such as delegation, memories, tool-calling, compaction, structured outputs, context engineering, and prompt design.
Strong scientific background, including comfort running experiments and understanding AI fundamentals such as dataset curation, training an AI system, designing evaluations and metrics, error analysis, and drawing sound conclusions from noisy results.

Nice To Haves

Experience building AI systems or agentic systems.
A degree in computer science, engineering, or math, or equivalent technical experience.
Experience with healthcare data or systems, such as EHRs, clinical documentation, or HL7/FHIR.
Experience taking an AI system from prototype to production at scale.
Published research or open-source contributions in machine learning or AI agents.