RAG/AI Developer

Project: building a RAG (retrieval-augmented generation) assistant on top of a local LLM stack, answering users with source-cited responses based on a knowledge base.

Responsibilities

• Building the ingest pipeline: document processing, chunking, metadata

• Setting up vector storage and embeddings (e.g. pgvector/PostgreSQL)

• Implementing retrieval and answer generation with cited sources

• Implementing and testing specific user use cases for the agent

• Integration with a local LLM server (Ollama) and multi-user interface (OpenWebUI)

Requirements

• Hands-on commercial experience building RAG systems/AI agents

• Experience with vector databases, embeddings, LLM frameworks (LangChain or similar)

• English - B2 or above

Nice to have

• Experience administering/hardening Linux servers - the project has adjacent tasks in this area

• Experience with Ollama/OpenWebUI

• German language

Conditions

• Project-based work, fully remote

• Priority given to candidates based in Eastern Europe

• Compensation: negotiable