Stefan Fuchs

Dr. Stefan Fuchs is a researcher at the Technical University of Munich (TUM), working in the Chair of Computing in Civil and Building Engineering within the TUM School of Engineering and Design. His research focuses on applying artificial intelligence techniques to challenges in the built environment, particularly in the areas of building regulations and compliance checking.

Dr. Fuchs completed his doctoral dissertation in 2024 titled "Transformer-based Semantic Parsing of Building Regulations: Supporting Regulators in Drafting Machine-Interpretable Rules" at ResearchSpace@Auckland. His work bridges the gap between natural language processing and construction engineering, developing novel approaches to interpret and represent building regulations in machine-readable formats.

His primary research interests include Natural Language Processing for building regulations, Automated Compliance Checking systems, Knowledge Representation and Reasoning techniques, and AI applications in the Built Environment. He has pioneered approaches using transformer models and intermediate representations to improve the semantic parsing of building regulations, enabling more effective automated compliance checking systems. His research spans both theoretical frameworks and practical tool development, including a LegalRuleML editor with transformer-based autocompletion.

Dr. Fuchs' publication record demonstrates consistent focus on transforming natural language building regulations into machine-interpretable formats. His work shows progression from foundational literature reviews to sophisticated transformer-based approaches, with recent publications exploring parallel drafting processes and large language model applications. This research trajectory indicates growing sophistication in addressing the complex challenges of regulatory compliance in construction.

He actively contributes to the academic community through teaching "Artificial Intelligence in Engineering" and maintains a strong research presence with profiles on GitHub, LinkedIn, ORCID, Google Scholar, and ResearchGate. His GitHub profile (stefan-1992) shows additional contributions to machine learning projects beyond his primary research focus, including work on "grokfast" related to accelerating machine learning phenomena.

Dr. Fuchs collaborates extensively with researchers including Prof. André Borrmann, Judith Fauth, and Robert Amor. His work is situated within TUM's broader research initiatives in Building Information Modeling, Digital Twinning, and Knowledge Representation and Reasoning, contributing to the university's leadership in construction informatics.