Doctolib is seeking a Staff Machine Learning Engineer - Retrieval to join our AI team and contribute to the Medical Knowledge Platform that grounds clinical AI decisions in validated medical sources such as HAS guidelines, learned society recommendations, and peer‑reviewed clinical studies. You will design and own indexing pipelines and retrieval systems, operate at scale (100M+ documents, 50+ requests per second, sub-300ms latency), and build custom re-rankers and robust query processing pipelines. You will engineer deep Retrieval‑Augmented Generation (RAG) systems that go beyond off‑the‑shelf components and work across the full retrieval stack: vector embeddings, vector search, re‑ranking, and query rewriting. Candidates should have proven production experience building and scaling search and retrieval in high‑traffic environments, with deep expertise in offline indexing and online retrieval and the ability to own the full pipeline. Hands‑on experience with Elasticsearch, Solr, Vertex AI, vector search, embeddings, re‑ranking, and query processing is expected, as is experience operating at Senior or Staff level with autonomy. Nice‑to‑have: experience with medical knowledge sources or healthcare information systems, evaluation frameworks for retrieval quality or AI outputs in regulated environments, and multilingual retrieval or domain ontologies. The role is based in Paris, France, with hybrid arrangements (up to two remote days per week) as part of Doctolib's cloud‑native tech platform. You will join a tech stack including Rails, TypeScript, Java, Python, Kotlin, Swift,

Doctolib is seeking a Staff Machine Learning Engineer - Retrieval to join our AI team and contribute to the Medical Knowledge Platform that grounds clinical AI decisions in validated medical sources such as HAS guidelines, learned society recommendations, and peer‑reviewed clinical studies. You will design and own indexing pipelines and retrieval systems, operate at scale (100M+ documents, 50+ requests per second, sub-300ms latency), and build custom re-rankers and robust query processing pipelines. You will engineer deep Retrieval‑Augmented Generation (RAG) systems that go beyond off‑the‑shelf components and work across the full retrieval stack: vector embeddings, vector search, re‑ranking, and query rewriting. Candidates should have proven production experience building and scaling search and retrieval in high‑traffic environments, with deep expertise in offline indexing and online retrieval and the ability to own the full pipeline. Hands‑on experience with Elasticsearch, Solr, Vertex AI, vector search, embeddings, re‑ranking, and query processing is expected, as is experience operating at Senior or Staff level with autonomy. Nice‑to‑have: experience with medical knowledge sources or healthcare information systems, evaluation frameworks for retrieval quality or AI outputs in regulated environments, and multilingual retrieval or domain ontologies. The role is based in Paris, France, with hybrid arrangements (up to two remote days per week) as part of Doctolib's cloud‑native tech platform. You will join a tech stack including Rails, TypeScript, Java, Python, Kotlin, Swift,