Geforderte Skills
- Retrieval-Augmented Generation
- Hybrid Search
- BM25
- Vector Search
- Reciprocal Rank Fusion
- Reranking
- Text Embeddings
- Chunking
- Search Indexing
- XML
- HTML
Nice to have:Information Retrieval EvaluationCode ReviewTechnical Leadership
Sprachen: French
Jobbeschreibung
Seeking an Expert Data & Retrieval (RAG) to join an AI Service team for a leading legal publisher. The role focuses on the upstream RAG pipeline, from content ingestion to result relevance.
Project overview
For a leading legal publisher, we are looking for an Expert Data & Retrieval to join the "AI Service" team. The team builds generative AI services that rely on a very large corpus of legal and editorial content. You will be responsible for the entire upstream part of the RAG chain: from content ingestion to result relevance.
Your missions:
- Design and industrialize the ingestion of editorial content (XML, PDF, HTML): parsing, normalization, enrichment with metadata.
- Define chunking, embedding, and large-scale indexing strategies, with incremental re-indexing and index versioning.
- Implement hybrid search (BM25 + vector), result fusion (RRF), and reranking.
- Measure and improve relevance (recall@K, nDCG, MRR, evaluation datasets).
- Manage the data lifecycle: updates, deletions, consistency between sources and indexes.
- Act as a technical referent: architecture choices, code reviews, team support.
Project details
- Start: flexible
- Remote: hybrid
- Location: 92400 Courbevoie, France, FR
- Languages: French
Indexiert von der 9am-Meta-Suche. Du bist Job-Anbieter? Listing beanspruchen oder verwalten
Gefunden auf collective.work