We are looking for a senior engineer or a small team to help us build an insurance distribution layer powered by LLMs and robust retrieval‑augmented generation (RAG). The focus is on structuring and querying complex insurance documents and exposing clean, well‑documented APIs. Responsibilities: Design and implement data ingestion pipelines for policies, quotes, endorsements, and claims. Build normalization and ontology mapping for coverages, exclusions, and limits. Implement a RAG architecture for accurate, explainable QA over insurance documents. Design and maintain OpenAPI‑documented endpoints for internal and partner use. Implement safeguards for regulated workflows, auditability, and traceability of model outputs. Requirements: Proven, production‑grade experience with LLMs and RAG (please share links/examples). Strong background with NLP for long documents (insurance/legal/financial preferred). Solid API engineering skills (REST, OpenAPI, auth, versioning). Experience with compliance/audit requirements in regulated environments. Ability to propose and implement an ontology/normalization layer over heterogeneous data sources. Nice to have: Direct insurance domain experience (broker/insurer platforms, policy admin, quote/bind). Experience with schema.org / structured data and SEO for AI answer engines. Experience with multi‑country compliance rule engines. Project details we’d like from you in your proposal: Links or descriptions for 1–3 relevant projects (RAG, doc parsing, regulated domains, or API platforms). A short architecture outline for: ingestion → ontology → exclusions → retrieval → APIs. Your approach to reducing hallucinations and ensuring answer citations. Estimated timeline and team composition for an MVP.