Mission: Build RAG System for Endangered Spoken Language

A job posting seeks an experienced NLP or LLM engineer to develop the first Retrieval-Augmented Generation (RAG) localization engine for a low-resource language spoken in South America. The project utilizes a proprietary corpus of pedagogical content and dictionary data developed over four years.

• Develop a RAG pipeline using LangChain or LlamaIndex with multilingual-e5 embeddings.
• Implement vector database solutions (Pinecone, Weaviate, or Supabase pgvector) with latency under 500ms.
• Create a modular prompt layer supporting six use-case templates including translation and localization.
• Build multi-tenant B2B SaaS infrastructure with strict data isolation, JWT auth, and configurable quotas.
• Deliver REST API with Swagger documentation, an admin interface for glossary management, and an offline SQLite bundle for React Native.

The role requires full intellectual property transfer to the client and is budgeted between 5,000–10,000€ for a 10-week duration.