AI Engineer: RAG & LLM Production Systems

expert.ai in Modena, Italy seeks an AI Engineer with 2–4 years of software experience to ship LLM-powered systems. You will design end-to-end RAG pipelines, implement embedding strategies, vector retrieval, and response generation, and work with multi-step reasoning.

You will use LangChain, LangGraph, LlamaIndex or custom orchestration, maintain MCP servers, optimize retrieval quality, and contribute to fine-tuning when needed. Strong Python and production-grade coding sense required.

Design and implement RAG pipelines end-to-end: ingestion, chunking strategies, embedding models, vector retrieval, reranking, and response generation.
Build agentic workflows using LangChain, LangGraph, LlamaIndex, or custom orchestration — including tool use, memory management, and multi-step reasoning.
Integrate and prompt-engineer LLMs for domain-specific tasks; contribute to fine-tuning efforts when needed.
Develop MCP servers and clients to standardize tool and context exposure across AI systems.
Maintain vector databases (FAISS, Pinecone, Weaviate, Qdrant, pgvector) and optimize retrieval quality.
Build evaluation pipelines to track hallucination rate, retrieval precision, latency, and output consistency over time.
Write clean, tested, production-ready Python and contribute to code reviews.
Collaborate with senior engineers and clients to translate requirements into solid technical decisions.

2 to 4 years of software engineering experience.
1–2 years focused on LLM or applied AI systems.
At least one production RAG or agentic application.
Solid understanding of embeddings, transformer fundamentals, context management, and prompt design patterns.
Familiarity with Docker.