TUM IDP: LLM Tooling Framework
Oct 2024 – May 2025 • TUM Interdisciplinary Project
An interdisciplinary research and software engineering project at TUM focused on automating the translation of research-stage software prototypes into production-grade software tools using Large Language Models.
Technical Scope
- Research to Production Framework: Designed a pipeline in Python that reduces manual engineering required to take academic/research tools to professional software standards.
- LLM Integration: Utilized Google Gemini and custom prompt orchestration workflows to analyze source code, refactor legacy code bases, generate unit tests, and produce developer documentation automatically.
- Software Engineering Standards: Applied automated validation steps, code quality checks, and structured test generation to ensure refactored components meet industry reliability standards.
Links and Resources
- GitHub Repository: github.com/carloslme/tum-idp