Senior AI Engineer — Health Search & Recommendations

Doctolib (France); France; Sur place; EUR 90 000 - 140 000; Plein temps

Avantages offerts par ce poste

Health insurance

25 days vacation + RTT

Lunch vouchers

Transport subsidy

Flexible days abroad

Résumé du poste

Doctolib is seeking a Senior AI Engineer to join the Patient team in Paris. You will design production search and recommendation engines powering a health companion used by millions, focusing on fast, relevant, and trustworthy results.

You will collaborate with ML Engineers, PMs and SWE to ship AI-powered features, focusing on embeddings, vector search, and RAG techniques, with strong emphasis on latency, reliability, and observability.

Qualifications

• Production deployment of algorithms to a service (ECS-based) on AWS.
• Strong analytical mindset with patient-first approach.
• Experience shipping search or recommendation systems to production.
• Hands-on end-to-end retrieval: ranking, reranking, embeddings & vector search.
• Familiar with RAG, LLM-based or managed rerankers.

Responsabilités

• Design and build production search and recommendation architecture: full retrieval, ranking, reranking pipeline.
• Establish strong baselines (prompts, evaluation, model selection) before custom ML.
• Build evaluation and observability into every stage, with offline and online evaluation.
• Collaborate with ML Engineers, PMs and SWE to ship AI-powered features delivering value.

Connaissances

Production deployment

Analytical mindset

Search/Recommendation systems

Embeddings & vector search

RAG / LLM-based rerankers

Architecture-first approach

Evaluation & observability

Low-latency systems

Outils

AWS ECS

Vertex AI

Python

TypeScript/React Native

Description du poste

Doctolib is seeking a Senior AI Engineer to join the Patient team in Paris. You will design production search and recommendation engines powering a health companion used by millions, focusing on fast, relevant, and trustworthy results.

You will collaborate with ML Engineers, PMs and SWE to ship AI-powered features, focusing on embeddings, vector search, and RAG techniques, with strong emphasis on latency, reliability, and observability.