Work with rich longitudinal signals from wearables plus real-world context.
Help turn ambiguous health questions into structured, testable AI systems.
Balance scientific depth, engineering pragmatism, and product impact so millions of members get guidance that respects both their biology and their real lives.
Design and build LLM‑backed product capabilities: Ship user-facing features that use LLMs and other AI models to deliver personalized insights, guidance, and proactive notifications. Implement safe tool-calling, retrieval, and orchestration so that AI components behave deterministically where they must and adaptively where they can.
Own evaluation, quality, and safety for AI workflows: Lead the design and implementation of evaluation frameworks and tooling to measure quality, safety, latency, and cost before and after release. Define the metrics and slices that matter for user-facing guidance, and integrate evals into the production pipeline.
Integrate LLMs with personalization and understanding layers: Ground AI behavior in structured user context rather than one-off prompts. Connect AI components to navigation flows, product surfaces, and action systems so guidance turns into coherent, multi-step programs and one-tap actions, not isolated tips.
Contribute to a multi-LLM and reasoning platform: Prototype and productionize workflows across multiple model providers and configurations, including routing logic and shadow-mode experimentation. Collaborate with infrastructure and science teams on reasoning, planning, and multimodal use cases.
Build robust, observable, and cost-aware systems: Design and implement services, workflows, and full stack product capabilities that meet reliability and performance expectations. Take ownership of operational health: debugging production issues, reducing technical debt, and iterating on architecture as the AI surface area and traffic grow.
Partner cross-functionally: Work closely with product, data science, research, design, and content to shape problem definitions, constraints, and evaluation plans. Communicate trade-offs clearly and help the team make principled decisions in a fast-moving domain.
Requirements
2+ years of hands-on experience in AI engineering, with a multi-year background in backend engineering, full stack product engineering, applied ML, or related roles building production systems.
Strong proficiency building production systems across the stack, including user-facing product surfaces and backend services, and comfort working with cloud-native services to ship and maintain production features.
Demonstrated ability to own systems end-to-end: from problem framing and data pipelines through modeling and prompting, all the way to deployment, monitoring, iteration, and product delivery.
A track record of working in product-facing teams, shipping to real users rather than only research prototypes, and caring about impact and iteration speed.
Comfort operating in a fast-changing AI/LLM domain with ambiguity, balancing rigor with pragmatism and keeping member value and safety at the center.
Excellent communication and collaboration skills, including the ability to explain complex technical trade-offs to non-technical stakeholders and work effectively in cross-functional teams across time zones.