About the RoleYou will own the patient facing AI conversation loop end to end, building robust agent systems that improve patient outcomes. You will work across tool integrations, evaluation frameworks, retrieval, and clinic side automation to ship code that directly impacts care quality.
What You'll Do- Diagnose why AI conversations with patients fail and ship fixes to prevent recurrence.
- Build and maintain agent infrastructure including tool integrations, evals, and retrieval systems.
- Automate recurring manual workflows on the clinic side to reduce toil through code.
- Handle patient facing interactions with warmth when the AI falls short and translate these moments into product insights.
- Own systems end to end with minimal oversight and make architectural decisions independently.
What We're Looking For- Experience building and shipping LLM or agent systems end to end across professional roles, personal projects, research, or open source.
- Hands on experience with agent infrastructure: evals, retrieval, and tool integrations.
- Strong CS and math fundamentals; elite academic signals are a plus.
- Automation mindset with a track record of removing recurring manual workflows via code.
- Warmth and empathy for patient facing interactions and ability to extract product learnings from conversations.
- Autonomy to own systems in a fast moving environment with minimal supervision.
- Existing US work authorization; visa sponsorship not available.
- Proficiency in Python or equivalent for building agent infrastructure.
Compensation & BenefitsSalary range: USD 150,000 to 220,000 annually.
LocationOn site in San Francisco, California, United States. Primary location: San Francisco, CA. Local remote options are not available for this role.
