Future - Staff Applied AI Engineer

Role overview Build the AI capabilities behind a health coaching product, turning language model workflows into useful, dependable features for members. The work covers agent design, integrations, evaluation, deployment, and ongoing improvements to response quality, speed, and cost, in collaboration with a cross-functional product and engineering team.

Responsibilities - Build and ship production agents using tool calling, structured outputs, parallel API orchestration, and streaming. - Develop features such as coach matching and guidance related to sleep, nutrition, and fitness, with human review where appropriate. - Run the development loop from prototyping and evaluation through deployment, trace monitoring, and iteration. - Improve reliability, latency, and cost through techniques such as caching, token limits, retries, and observability. - Create validated agent tools and integrations as part of a cross-functional product team.

Requirements - 4–10 years of experience, including having shipped LLM agents into production. - Strong Python and asynchronous programming skills, with experience building agent workflows. - Ability to work independently in a beta-stage product environment. - US citizenship or permanent residency; visa sponsorship is not available.

Nice to have - Experience with LangChain or LangGraph, Pydantic, JSON Schema, and LLM evaluation and monitoring tools. - Familiarity with AWS or Bedrock, OpenTelemetry, Datadog, Terraform, or CDK. - Background in a large technology company followed by startup experience, or experience founding a company.

Benefits and work setup - Fully remote within US time zones, with optional team weeks at the Los Angeles headquarters. - Base salary of $250,000–$300,000 for US candidates, plus equity.