Senior Software Engineer

Marvin Behavioral Health helps people navigate their mental health and wellbeing, partnering with employers and health plans to give members one place to understand coverage, find care, and get support. You will join a small team shipping a production healthcare application where every engineer owns work end to end across product and infrastructure. AI is woven into both the member experience and the engineering workflow, supporting faster, higher-quality delivery without compromising reliability.

This Senior full-stack role is designed for someone who builds with AI as a core engineering capability. It is not a machine learning research position: you will not train models or tune GPUs. Instead, you will ship product in a modern TypeScript and Next.js stack, while creating LLM-powered experiences and evaluation practices that keep AI features dependable with sensitive health data.

What you’ll do

• Own features end to end across the Next.js/React/TypeScript app, Prisma/Postgres backend, and AWS infrastructure.
• Extend Marvin’s AI coach and build new LLM-powered member experiences, including retrieval over benefits data, structured extraction, guided flows, and agentic tool use against your own APIs.
• Design evaluation and observability for AI features so regressions are detected when models change, not only when failures crash.
• Advance AI-assisted development by building and refining agent instructions, reusable skills, hooks, MCP integrations, and review practices that raise quality as velocity increases.
• Handle PHI responsibly with clear boundaries, redaction, audit trails, and vendor data agreements treated as design requirements.
• Review code critically whether it comes from humans or agents, and mentor teammates while improving testing and reliability.

What you’ll bring

• 5+ years building production web applications with deep TypeScript, React, and Node experience.
• Next.js is a strong plus.
• Comfort across the stack, including relational data modeling (Prisma and Postgres), auth flows, caching, background jobs, and cloud deployment (AWS, Docker, Terraform or similar).
• A strong testing habit at every layer: unit, component, and end-to-end.
• Hands-on with Vitest, Storybook, Cypress, and Playwright.
• Strong product judgment and ability to work directly with design, clinical, and partner-facing teammates.
• Daily expert use of agentic coding tools such as Claude Code, Cursor, or Codex.
• Hands-on experience shipping LLM features: prompt design and versioning, structured outputs, tool/function calling, streaming UIs, retrieval-augmented generation, and multi-turn conversation state.
• Experience running evals and tracing production LLM traffic using LangSmith, Braintrust, Langfuse, or similar, with an ability to explain changes after model version bumps.
• Practical understanding of the cost, latency, quality trade-offs, including model selection, caching, prompt compression, and fallback strategies.
• Clear-eyed constraints: knowing when an LLM is the wrong tool, how to constrain when it is the right tool, and how to prevent hallucinations from reaching members.

Nice to have

• Healthcare, benefits, or other regulated-data experience (for example HIPAA or SOC 2).
• Building MCP servers or internal agent tooling for a team.
• Experience with Clerk, Stripe, or Sentry.
• Contributions to AI developer tooling, open source, or internal projects.

Compensation & benefits

• Base salary: USD 160,000 - 190,000 per year
• Equity
• Health, dental, and vision
• [401(k)]
• Unlimited PTO
• Hybrid: 2 days/week in office (Los Angeles or Denver)

Location: Los Angeles, CA (hybrid). Seniority: Senior. Experience level: 5+ years.