Work on the technical foundation for deploying Large Language Models and embedding models in our enterprise search and AI platform
Evaluate new LLMs, embedding models and, prospectively, multimodal models (e.g., vision)
Perform qualitative and reproducible assessments of model quality in the context of real enterprise data (RAG scenarios)
Evaluate, integrate and update LLM operational and inference components (e.g., vLLM, TEI, LangChain or comparable frameworks)
Maintain and further develop the technical LLM platform (compatibility, stability, updates)
Develop internal tools and utilities for model, prompt and quality evaluation
Perform prompt engineering and prompt tuning for different models and use cases
Evaluate and implement integration and interface concepts (e.g., MCP or similar standards)
Documentation, testing and internal technical support for other development and product teams

Enthusiasm for building and operating technical platforms in the areas of AI / LLM / RAG
University degree in (business) computer science, mathematics, natural sciences or a comparable field
Practical experience with Large Language Models and/or embedding models
Experience with Retrieval-Augmented Generation (RAG)
Experience with LLM frameworks, inference or serving components
Confident in Python and experienced in developing robust software components
Good understanding of quality, traceability/explainability and the limitations of AI systems
Structured, independent and responsible working style and strong team skills
Excellent German and English skills, spoken and written (native German speaker or at least level C)

AI Platform Engineer, LLM, RAG

Key skills