Early-stage healthcare AI startup

Platform Engineer

Apply
Full-timeGreater Los AngelesHybrid$150,000 to $300,000

Company Description

An early-stage, small healthcare technology company building AI-driven automation for operational work. The team is AI-native and growing quickly, and AI agents already do a large share of its day-to-day engineering.

Because the company works with sensitive health data, reliability, security and careful engineering are central to how it builds.

Job Description

This is an infrastructure role with an unusual twist: many of the people using what you build are AI agents, alongside the engineers who supervise them. You will not write the company's automations yourself. You will build and run the platform that lets agents investigate problems, write and review code, ship changes and confirm they worked in production.

You will own the environment agents run in, the pipeline their changes go through, the observability that shows what is happening across the system, and the guardrails that keep an unchecked change from reaching customers. The work spans distributed systems, durable execution, deployment and versioning, reliability and security.

The codebase is strictly typed TypeScript written in a functional style, with the compiler doing much of the work of keeping changes correct. You will work closely with engineering leadership to set the standards the team, human and agent alike, builds on.

You will:

  • Stand up and run secure, isolated environments where AI agents do their work, with tightly scoped access to secrets and outside services
  • Shape the path a fix travels, from first report to a verified production change, including code review and approval steps
  • Make releases quick and repeatable, with versioning and rollback that work without manual effort
  • Give the team visibility into the platform through metrics, alerts and tracing tied back to specific builds
  • Design command-line tools, APIs and error output that agents can act on and troubleshoot on their own
  • Keep long-lived distributed processes correct and able to recover when parts of the system fail
  • Create review and approval tooling that lets engineers oversee and steer agent activity
  • Respond to production incidents, lead postmortems and address underlying causes

Job Requirements:

  • Several years of building and operating systems in production
  • Strong grounding in distributed systems concepts such as retries, idempotency, concurrency, backpressure and partial failure
  • A track record of owning reliability for a critical system, including on-call duty and debugging with metrics, logs and traces
  • Practical cloud experience (AWS, GCP or Azure) covering CI/CD, containers and safe release practices
  • Strong skills in TypeScript, which is the primary language here; excellent engineers coming from Go, Rust or a comparable typed language are also considered
  • Regular, hands-on use of AI coding tools and a clear sense of where they fall short
  • Willingness to work across the stack, mainly backend and infrastructure with occasional frontend
  • A strong bar for simple, readable, maintainable code
  • Functional programming background, e.g. Haskell, OCaml, F#, Rust or Scala (nice to have)
  • Familiarity with durable workflow engines or similar orchestration frameworks (nice to have)
  • Experience with internal developer platforms or build and release tooling (nice to have)
  • Experience operating autoscaled container or VM fleets and observability stacks such as Prometheus, Grafana or OpenTelemetry (nice to have)
  • Experience shipping LLM-based agents to production (nice to have)
  • Prior work in a compliance-heavy industry where data security is critical (nice to have)

Apply for this Position