Forward Deployed Engineer (Mid/Senior)

About the Role Role Mission As HAI's Forward Deployed Engineer, you will be the technical owner of AI deployments that directly transform how health systems operate. You'll embed with customers to build, launch, and operate production conversational AI agents—architecting systems that handle real clinical workflows and impact thousands of patient interactions. This role exists because healthcare organizations are ready to deploy breakthrough AI at scale, and they need a world-class AI engineer who can build reliable, innovative systems in the field. What You Will Accomplish Own your first major outcome: By day 90, you will have completed end-to-end ownership of your first AI deployment: designed and implemented a RAG pipeline grounded in customer data, built tool-calling and MCP integrations connecting our agents to customer systems (EHRs, data warehouses, operational tools), executed a production go-live with zero surprises, and established monitoring that catches anomalies before customers do. Drive lasting impact: At 12 months, you will have deployed multiple agents across your assigned health system, built reusable AI patterns and frameworks that accelerate future deployments, become the trusted technical partner that customers rely on to solve their hardest AI problems, and generate measurable evidence that our agents improve operational reliability and clinical outco [... source excerpt omitted ...] ealthcare environments. The Team You'll work alongside Deployment Strategists, engineers, and clinical experts—embedded with customers but tightly connected to our core AI team. You'll operate with high technical ownership and autonomy in the field, with direct access to our product, ML research, and engineering leadership. This is a culture of shipping real systems, owning outcomes, and solving problems before they become crises. What You Will Do Design and implement RAG pipelines that ground conversational AI responses in customer clinical data, ensuring accuracy, safety, and relevance to healthcare workflows while managing retrieval latency and data governance Build too [... source excerpt omitted ...] to interact securely with customer systems—EHRs (Epic, Cerner, Athena), data warehouses, and operational tools—handling errors gracefully and enforcing safety constraints Develop production Python code using LangChain, LangSmith, and modern AI frameworks to implement advanced LLM techniques (RAG, prompt engineering, LLM-as-judge, chain-of-thought reasoning) solving novel healthcare AI problems Execute end-to-end deployments including infrastructure setup, integration testing, production monitoring configuration, cutover planning, and go-live execution—ensuring deployments happen on schedule without surprises Monitor and own production systems by instrumenting deployed agents, re