We are developing a web-based internal tool to streamline research project management. The tool will leverage Large Language Models (LLMs) to automate key tasks such as document review, compliance checking, and reporting for funded projects. It will incorporate retrieval-augmented generation (RAG) pipelines to provide intelligent, context-aware support for grant tracking and research administration.

Responsibilities

• Build and maintain a minimal, maintainable backend to support LLM-powered features
• Integrate LLM APIs (e.g. OpenAI, Hugging Face) for summarization, compliance checking, and Q&A
• Implement RAG pipelines for document ingestion, embedding generation, and semantic search (e.g. LangChain, LlamaIndex)
• Handle document parsing (PDF, DOCX)
• Expose RESTful APIs for use by a frontend interface
• Support simple internal deployment workflows (e.g. via Docker)

Required Skills

• Strong Python programming skills
• Experience with backend frameworks like FastAPI or Flask
• Familiarity with LLM APIs and prompt engineering fundamentals
• Experience with vector databases and semantic search (e.g. FAISS, Qdrant), including RAG concepts
• Comfortable with Git-based version control
• Proficiency in written and spoken English

Nice to Have

• Interest or experience in frontend development (e.g. Streamlit, Gradio, React)
• Basic understanding of DevOps for deployment (e.g. Docker)

We Offer

• Flexible working hours and remote work options
• Hands-on experience with in-demand AI technologies, including LLMs and RAG pipelines
• A practical, applied project with real-world impact in academic research administration
• TUM HiWi contract of up to 20 hours per week

Interested?

Please send your complete application, including your CV and letter of motivation, preferably in one PDF file, by July 23, 2025 to Lukasz Mazur at lukasz.mazur@tum.de with the subject line:
TUM AIR Student Assistant for LLM Tool Development Application.

TUM has been pursuing the strategic goal of substantially increasing the diversity of its staff. As an equal opportunity and affirmative action employer, TUM explicitly encourages nominations of and applications from women as well as from all others who would bring additional diversity dimensions to the university's research and teaching strategies. Preference will be given to disabled candidates with essentially the same qualifications. International candidates are highly encouraged to apply.