Senior AI Data Engineer– Agentic Healthcare Platform

The Role This is a full-ownership data engineering role at the center of Medeloop's AI platform. You won't be maintaining pipelines someone else built, you'll be architecting the data backbone that powers AI agents doing real operations at scale. You'll work directly with data scientists, AI engineers, and product teams to turn raw, complex healthcare data into the clean, structured, semantically-rich foundation our AI scientists depend on.

Candidates who currently perform these tasks exclusively through manual processes are unlikely to be suitable for this role. If you want to build something that genuinely changes how medical research gets done, this is the role. What You'll Own The healthcare data lake: curating, extending, and evolving it through new concepts, derived variables, and data models that directly inform our AI engines and customer products AI-native data workflows: designing and operating AI-powered pipelines (using tools like Claude Code and agent frameworks) to automate harmonization, cleaning, quality checks, and summarization at scale NLP and semantic infrastructure: building pipelines for entity extraction, concept normalization, embedding-based retrieval, and semantic search that power the AI Scientist platform Novel data extraction approaches: experimenting with and building new methodologies for working with unstructured clinical data, not just applying existing playbooks Research-grade data products: delivering analytical samples, cohorts, and final datasets that withstand scientific scrutiny and are actively used by researchers and customers Data governance and observability protocols : including access controls, PHI/PII handling, data classification, compliance, monitoring, alerting, data freshness, and comprehensive documentation to enable self-service capabilities.

What We'Re Looking For

3+ years of relevant data engineering or data management within an analytics-driven organization, with end-to-end ownership from raw ingestion to final data product Deep hands-on experience with healthcare CDMs (OMOP, FHIR, PCORnet) — designing or extending them, not just querying Knowledge of medical ontologies: UMLS, SNOMED CT, RxNorm Experience with big data, data pipelines and tooling that support retrieval-augmented generation (RAG), vector integrations, embedding workflows, and other AI/ML workloads.