WhyHireWrong?

Compensation

Full Description

About the company This is a pre-seed, well-funded stealth healthcare AI company. Its founders have taken two healthtech companies to unicorn status. They include the former VP of Engineering at Tumblr and the founding CTO of Oscar Health. The company is building autonomous agents that carry out full doctor tasks inside large healthcare organizations. This is not a chatbot or a summarizer. The agents take on clinical and administrative work that a physician does today, so their reliability matters a great deal. Trials are already running, and more are coming. Why this role exists The engineering team has 11 people, with backgrounds from Google Zurich, Oxford, TUM and NVIDIA. The company is hiring an individual contributor to build and own the core agent systems from scratch. This is not a team lead role. The Staff / Principal title shows how much ownership and maturity the role carries. You will write code every day, and your decisions will shape the architecture for years. What you will do Design and build the agent harness: how agents plan, call tools, recover from failure and hand off between steps. Build multi turn routing so each conversation and task reaches the right agent, tool and model with the right context. Integrate tools and external systems through MCP and custom tool interfaces, and make those integrations safe and testable. Own prompt management, versioning and evaluation, so every change to agent behavior is measured before it reaches a hospital. Build the distributed backend the agents run on, including queues, state, retries, observability and latency control. Design audit trails and access controls so every agent action can be explained and reviewed. Healthcare compliance such as HIPAA shapes the design from the first day. Work directly with the founders and the first healthcare customers. You will see how agents behave in real clinical settings and fix what breaks. What you need You currently hold, or recently held, a Principal, Staff or Senior title in ML, AI or software engineering. Your recent work includes writing production code. You have built software from zero to one at a startup, from first commit to real users. You have shipped AI agents, tool calling or multi turn routing systems to production. You can explain the failures you hit and how you fixed them. You write production quality Python and can work comfortably in at least one of Go, Rust or TypeScript. You have built and operated distributed systems at scale, and you have a clear view on state, concurrency and failure handling. You have hands on experience with LLM APIs, evaluation of non deterministic systems, and prompt and context management. You are based in Europe or EMEA and can work with the team across European time zones. The company cares most about intellectual horsepower, sound engineering principles and a record of shipping. Your degree does not need to be in a specific subject, though many engineers on the team studied Math, Physics or CS. Experience at top AI labs, AI native startups and large scale product companies is typical across the team. Feedback after each stage is quick, and next steps are usually decided within about 3 days. Before applying: Work history and identity are confirmed before the first screen.