Diese Seite auf Deutsch ansehen

Prof. Dr. Hartmut Westenberger

Dr. rer. nat.
Faculty of Computer Science and Engineering Science

Cologne Institute for Digital Ecosystems

Prof. Dr. Hartmut Westenberger

Campus Gummersbach
Steinmüllerallee 1
51643 Gummersbach
Room 2.231 Mailing address


  • Phone: +49 2261-8196-6385

Office hours

Public Office Hours & Consultation
Wednesday, 13:30 to 14:15
Campus Gummersbach, Room 2229/2231/2224
Public Hours should be booked by using the calendar functionality "consultation hours" of the ILIAS calendar "News für Studierende der F10" see ...... https://ilu.th-koeln.de/goto.php?target=grp_176826&client_id=thkilu .....

Positions

  • Institute Director
  • Program Manager of the international Master Program "Digital Sciences / Business Information Systems"; Director of the Cologne Institute for Digital Ecosystems

Official capacities

  • Examination affairs
  • Member of the THK-AI Research Cluster
    The THK-AI Research Cluster, under the guiding principle "Shaping social innovation with AI," serves as the central point of contact for all matters related to artificial intelligence at the university. It initiates and promotes collaborative projects between associations, industry partners, professors, and students.
    THK-AI Research Cluster

Preprint

  • Benchmarking Large Language Models for ABAP Code Generation: An Empirical Study on Iterative Improvement by Compiler Feedback
    Stephan Wallraven, Tim Köhne, Hartmut Westenberger, Andreas Moser, 21.01.2026, Publisher: arXiv
    This work investigates the performance of Large Language Models (LLMs) in generating ABAP code. Despite successful applications of generative AI in many programming languages, there are hardly any systematic analyses of ABAP code generation to date. The aim of the study is to empirically analyze to what extent various LLMs can generate syntactically correct and functional ABAP code, how effectively they use compiler feedback for iterative improvement, and which task types pose special challenges. For this purpose, a benchmark with 180 tasks is conducted, consisting of adapted HumanEval tasks and practical SAP scenarios. The results show significant performance differences between the models: more powerful LLMs achieve success rates of around 75% after several iterations and benefit greatly from compiler feedback, while smaller models perform significantly weaker. Overall, the study highlights the high potential of powerful LLMs for ABAP development processes, especially in iterative error correction.
    Benchmarking Large Language Models for ABAP Code Generation
  • Customized Generative AI – mature for operational usage?
    Digital EXchange, 17.09.2025, Gummersbach - The presentation explores the operational readiness of customized generative AI as it was im Sept 2025, focusing specifically on Retrieval-Augmented Generation (RAG). The authors illustrate its utility through a case study involving a support chatbot designed to navigate the complex web of university examination regulations. By comparing RAG with fine-tuning and prompt engineering, the sources highlight how hybrid AI systems can balance data accuracy with cost efficiency. The current market maturity of these Technologies is evaluated, noting that while expectations are high, successful deployment requires rigorous data governance and quality control. Ultimately, a framework for assessing AI maturity and the technical requirements are discussed, necessary for moving from experimental pilots to productive use.

Up
M
M