Senior Software Engineer / AI Engineer

9w2 months ago

Midi Health

US · Full-time · $170,000 – $210,000

About this role

Midi Health is the largest virtual care clinic for women in midlife navigating perimenopause, menopause, and other hormonal transitions. We combine expert clinicians, evidence-based protocols, and a modern technology platform to deliver underserved care. AI is reshaping clinical care, and you'll build products at the frontier of that shift.

You'll design, build, and ship LLM-powered features end-to-end, including prompt design, retrieval, tool use, agents, evaluation, and production operations. Build evaluation harnesses and feedback loops to ship AI features responsibly. Integrate AI deeply into patient, clinician, and operational workflows.

Hybrid by design with two days a week in-office (Tuesday and Thursday) in Palo Alto or San Francisco. AI-native engineering expects fluency with tools like Claude Code, Cursor, Copilot. Low-ego, high-curiosity teams approach disagreement with humility and invest in growth.

Outcome-oriented culture moves fast while balancing speed with healthcare durability. Partner with clinical, product, and safety stakeholders to define success. Stay close to AI research and ecosystem to bring frontier techniques to the team.

Requirements

6+ years of software engineering experience, with meaningful recent time building production LLM / ML features
Strong hands-on coding skills in Python; comfortable across the full stack when the feature demands it
Seasoned at system design for AI systems: retrieval, orchestration, caching, evaluation, cost and latency tradeoffs
Deep, hands-on command of modern AI coding tools — strong follower of the frontier and apply new techniques quickly
Good mentorship instincts; generous with what you learn
Rigorous about evaluation and safety; allergic to vibes-only launches
Strong communicator; comfortable explaining model behavior to non-technical stakeholders
Low-ego, curious, humble

Responsibilities

Design, build, and ship LLM-powered features end-to-end: prompt design, retrieval, tool use, agents, evaluation, and production operations
Build evaluation harnesses and feedback loops to ship AI features responsibly
Integrate AI capabilities deeply into patient, clinician, and operational workflows
Partner with clinical, product, and safety stakeholders to define what 'good' looks like
Contribute to internal AI-for-engineering practices and tooling
Stay close to the research and ecosystem; bring the best of it back to the team

Benefits

Hybrid by design: two days a week in-office (Tuesday and Thursday)
AI-native engineering: fluent with modern AI coding tools like Claude Code, Cursor, Copilot
Low-ego, high-curiosity: approach disagreement with humility and technical rigor
Outcome-oriented: move fast, own work end-to-end, balance speed with healthcare durability