Medical AI Data Engineer 80-100 %

Key Responsibilities : Design, develop, and maintain an AI agents leveraging RAG and autonomous AI agents, enabling clinical researchers to query sensitive federated datasets using natural language Build and iterate on an interoperable query language to SQL generation pipeline, including schema-awareness layers, and query validation tailored to complex clinical data models Implement and maintain data harmonization pipelines conforming to the OMOP Common Data Model and HL7 FHIR standard, enabling interoperability across heterogeneous clinical datasets from multiple institutions Integrate and maintain medical ontology mappings across ICD-10, SNOMED CT, ATC, and RxNorm to ensure semantic consistency in federated patient data queries Develop and maintain full-stack platform features: from backend services, APIs, and query orchestration pipelines through to user-facing interfaces for clinical researchers and data scientists. What We Are Looking For AI/LLM Systems: Hands-on experience building production RAG pipelines and agentic AI workflows applied to clinical data querying. Medical Ontologies: Practical experience integrating a broad range of clinical ontologies and vocabularies into a query or data harmonization system, including but not limited to SNOMED CT, ATC, RxNorm, ICD-10, LOINC, and OMOP standard vocabularies. The candidate must demonstrate hands-on implementation experience managing concept mappings, hierarchies, and cross-ontology relationships. Backend Engineering: Solid Go development experience, including API design and data pipeline construction. Frontend Engineering: React development experience, including component design and building secure, clinical-facing user interfaces. Strong SQL expertise: Writing and optimizing complex queries and investigating, monitoring, and troubleshooting the performance of production databases such as PostgreSQL and Oracle. Preferred qualifications Experience working with relational databases and complex clinical data models. Proven ability to build and maintain production React interfaces, ideally within a data-intensive or analytics-focused application. Prior experience working within privacy-preserving, federated or distributed data environments where data cannot be moved or revealed in raw form is strongly preferred. Understanding of the engineering constraints this imposes on query design and data access is essential. Given the complexity and proprietary nature of our platform, we place significant weight on candidates who have sustained deep contributions to a single, specialized system over an extended period (18 months or more). Why Join Us? Meaningful Work: Your contribution to the infrastructure enables secure access to insights of patient data and enables life-saving research. Lausanne Hub: Work in the center of Europe’s Health Valley. Impact: In a team of <20, you aren't just a ticket-solver. You are building the foundation the company stands on. Challenging Problems: If you enjoy working at the boundary of AI, regulated health data, and distributed systems, you will not run out of hard problems here. This role is right for you if you: Thrive working across the full stack: from ontology mappings and data models to APIs to user interfaces 0-2 years working within a privacy-preserving or federated health data environment. Have a strong instinct for data quality and semantic correctness, especially in a clinical or medical context Appreciate challenging yourself out of your comfort zone and enjoy change. Are comfortable taking responsibility and being accountable for outcomes. Stay humble and open to feedback, especially when working with world-class experts.