Founding Research Engineer

Description

The company is an AI longevity lab that helps people stop, and sometimes reverse, chronic disease and slow age-related decline, using AI and established behavioral science.

Proactive care like that of the $100K concierge clinics should not be limited to the wealthy. The company partners with healthcare providers to offer its AI and clinician platform to their patients, covered by insurance.

The founders include a physician, and the team has people from Meta Superintelligence Labs, Headspace, Levels Health and Stanford. Medicare's Innovation Center selected the company for its ACCESS program, which pays for health outcomes rather than services, so patient outcomes are the only thing the business optimizes for. Revenue has grown 10x in four months.

This is the first AI hire, to found the research lab. Most RL environments are simulated, but this one is built on real data, with a goal of millions of intervention and outcome pairs within a year. Each pair is tied to a patient's full medical history and tracked over months and years.

Responsibilities

• Start the lab and set its technical direction.
• Build the environment and the agents that learn from it.
• Own the agents, evals and learning loops that turn longitudinal patient data into action.
• Publish benchmarks and models.
• Help shape the team that grows around the lab.
• Answer the core question: given what is known about a patient, what should happen next, what can an agent safely do, and did it work?
• Work through the cycle of patient state, agent decision, intervention, outcome and a better agent, with clinicians involved and every improvement reaching real patients.
• Build systems that merge fragmented clinical, behavioral and conversational history into a model of each patient that keeps updating.
• Give agents context, memory, tools, verification and escalation paths so they act reliably over days and months.
• Create evals from real failures and use them to improve context, tools, orchestration, model choice and post-training.
• Link agent decisions to patient behavior, clinical follow-through and outcomes.
• Run systems at production scale with low latency and cost.

Requirements

• You have shipped frontier AI systems with real-world impact and can explain how they failed, how you measured it, what you changed and whether the change helped.
• You know the agent stack broadly: architecture, context engineering, memory, retrieval, tool use, model routing, evals, guardrails and observability.
• You have strong engineering fundamentals, own systems end to end, work without a detailed spec, and prefer simple experiments over elaborate architecture.
• You have 6+ years of experience.
• Skills: Python, Machine Learning, Deep Learning, Reinforcement learning (RL), Evals, AI Agents.
• Strong plus: post-training (fine-tuning, RL, reward modeling, outcome-based training) and early-stage or founding experience.
• Healthcare experience is not required.

Conditions

Base salary of $250K to $285K plus 0.5% to 1.0% equity.

San Francisco Bay Area preferred; remote work within the US is considered.

Only US citizens or people with a US visa are eligible.

Recruiters contact candidates only from the company's own email domain; be wary of anyone posing as a recruiter.