Senior Data Scientist, AI Retrieval Systems

Role overview Build the biomedical data and retrieval layer that supports AI tools for rare disease research. This senior individual contributor role spans knowledge modeling, search and ranking, service development, user interfaces, and deployment to on-premises and high-performance computing environments. You will work with research and clinical specialists to make results useful, reviewable, and grounded in evidence.

Responsibilities - Ingest and maintain disease and phenotype vocabularies in PostgreSQL, reconciling identifiers and clinically relevant terms across sources. - Develop retrieval-augmented services that map everyday language to clinical concepts using embeddings, candidate retrieval, and model-assisted disambiguation. - Tune keyword and vector search, and build relevance ranking that balances retrieval quality and response time, including for conditions with limited curated coverage. - Create user-facing flows and status displays with React, Next.js, and TypeScript. - Deploy and operate services in on-premises and HPC Kubernetes environments, coordinating with the teams responsible for that infrastructure. - Maintain evaluation and regression suites, capture decisions and operational details for review, and contribute to research manuscripts and presentations.

Requirements - A bachelor’s degree in a relevant computing, data, or biomedical field, or equivalent professional experience; an advanced degree is preferred. - At least 5 years building and operating production software or data systems, including 2 years shipping LLM-powered applications used by others. - End-to-end retrieval experience, from indexing and query design through quality measurement on real data. - Experience evaluating systems with no single correct answer using golden sets, regression suites, or measures such as Recall@K and MRR. - Experience constraining model outputs to typed schemas and validating results, plus ownership of services from data design through operations. - Ability to obtain and maintain a Public Trust security clearance.

Nice to have Experience with biomedical ontologies, clinical terminology, on-premises or HPC Kubernetes deployments, open-weight model serving, clinical genetics or rare disease research, NIH programs, or public biomedical standards and resources. Experience explaining technical work through publications, presentations, or open-source contributions is also valued.

Benefits and work setup The position is remote within the United States. Benefits include employee medical, dental, and vision coverage, paid time off and holidays, a 401(k) match, education support, referral bonuses, and flexible spending and transportation accounts.