AI Engineer (LLM & RAG) with n8n

Role overview Build AI assistants and agents that use a client's own documents and tools, with safeguards and evaluation that help make their output dependable. The work includes document-grounded question answering with citations, drafting and sorting tasks, and LLM features embedded in client products, with an emphasis on recognizing uncertainty instead of presenting unsupported answers as facts.

Responsibilities - Build retrieval pipelines over sources such as PDFs, shared drives, knowledge bases, and code repositories, using PostgreSQL and pgvector. - Design prompts and structured response formats for language models. - Create evaluation sets from real client data and run them to assess changes. - Set and monitor cost and latency limits; use caching and provider fallbacks where appropriate. - Explain implementation trade-offs to clients in clear, nontechnical language.

Requirements - At least two years of Python development in production. - Experience shipping an LLM feature used by real people. - Understanding of embeddings, document chunking, and retrieval quality. - Habit of measuring outcomes before and after making changes.