AI Product Engineer

Responsibilities

• Design and build AI agents and automated workflows for healthcare operations, including insurance benefits verification, prior authorizations, patient intake, claims, and clinical documentation.
• Build web and mobile features for clinicians, patients, families, and internal operations teams.
• Define ambiguous problems with clinicians and operators, write specifications, and take features from concept through first release.
• Design services, data models, and APIs, and manage infrastructure as code.
• Own production services, including deployments, monitoring, and incident response.
• Write evaluations for model-facing features and track workflow performance metrics.
• Handle protected health information using least-privilege access, auditable changes, and PHI-free logs.
• Build and maintain engineering skills, plugins, MCP integrations, and review bots for Claude Code and Codex.

Requirements

• 5+ years of experience building and shipping production software.
• Recent hands-on experience using or building AI harnesses involving project instructions, tools, MCP servers, skills, hooks, subagents, and evaluations.
• At least one personal project published on the web that is live and usable, preferably a SaaS product with working payments such as Stripe.
• Strong full-stack experience with Python and TypeScript across frontend, backend, databases, and cloud infrastructure.
• Experience owning products or features end to end and working in a large, established codebase.
• Hands-on experience with LLM tool use, agent orchestration, structured outputs, retrieval, and evaluations, using LangGraph, an agents SDK, or similar.
• Daily experience using and configuring AI coding agents such as Claude Code, Codex, or Cursor.
• Product judgment for making UI decisions without a dedicated designer and clear written communication in a distributed team.
• Preferred experience includes healthcare or regulated industries, HIPAA, payer and claims workflows, EHR integrations, HL7/FHIR, Temporal, React Native, production LLM evaluations or safety checks, agent harnesses, MCP servers, coding-agent plugins, open-source work, or widely used internal tools.