AI Product Engineer, Clinical AI

About the company

Our client is a seed-stage clinical AI company in New York City building an execution platform for risk-bearing care organizations. The product targets chronic condition management, with particular attention to rural and underserved patients, and is used by primary care and multispecialty groups across the United States. It has two sides: agents that ingest a patient's full chart and give clinicians specialist-level recommendations inside the EHR during the visit, and a patient-facing side where the company's own clinical team and agents execute parts of the care plan under the client's brand. Underneath both sits a longitudinal patient context graph, with condition-specific subagents handing off between the clinician and patient sides in real time. Founded in 2025, roughly $13M raised across a pre-seed and a seed, a team of about 12 to 21 drawn from leading value-based care and digital health organizations, and production results showing 3x faster chronic condition control.

The role

You'd own the agentic clinical platform end to end — context management, tool use, orchestration, permissions, memory, and human-in-the-loop workflows. This is not greenfield: the foundations already exist and your job is taking them to scale, adding governance on the eval platform, observability, and durable architecture. Day to day you'd prototype new clinical AI experiences quickly, then harden the ones that work into production software real clinicians depend on. You'd design the evals, regression suites, and launch criteria that decide when something is ready — task success, clinical quality, reliability, safety, latency, and cost — and build observability across model behavior, tool calls, traces, errors, and end-to-end outcomes. You'd also shadow the founders in customer meetings, talk to clinicians directly, and come back with strong opinions. The work is mostly backend, but you'd be expected to move across DevOps, data, and frontend. You'd report to the Head of Engineering and work side by side with one of the co-founders, on an engineering team of five core engineers plus design and informatics.

What you'll do

• Own the agentic clinical platform end to end: context management, tool use, orchestration, permissions, memory, and human-in-the-loop workflows
• Take existing foundations to scale — governance on the eval platform, observability, and architecture
• Prototype new clinical AI experiences fast, then harden the best ones into production software used by real clinicians
• Design evals, regression suites, and launch criteria across task success, clinical quality, reliability, safety, latency, and cost
• Build observability across model behavior, tool calls, traces, errors, and end-to-end outcomes
• Shadow the founders in customer meetings, talk to clinicians, and come back with strong opinions
• Move across the stack as needed — mostly backend, but into DevOps, data, and frontend

What we're looking for

Nice to have

• Founder or very early employee experience — joined at seed or Series A and helped scale
• Data engineering depth, which this client values strongly
• A modern React framework such as Next.js, GCP, and Postgres-compatible databases
• An active GitHub or side projects showing initiative
• Direct experience with clinical data, EHR integrations, or RAG over medical records

Tech stack

• Python and TypeScript (Go is a plus, not required)
• LLMs, agentic frameworks, and RAG
• Next.js
• GCP
• Postgres-compatible databases

Compensation & logistics

• $220k–$245k base + competitive equity at seed stage
• Hybrid 3 days/week in New York City — flexible days, occasionally 4 to 5
• Relocation support available
• Visa transfers only — OPT and H-1B transfers are OK, but there is no new sponsorship
• Reports to the Head of Engineering, working alongside a co-founder
• The process includes a 48-hour take-home presented at a half-day on-site in NYC
• Full-time · hiring 1 for this role