AI-assisted Softwaredesign and -engineering (KDE)

How far does a team get when AI is part of development from the very first idea? You will go through the entire path from concept through prototype to deployment — using local language models, your own coding harness, and a direct comparison with commercial systems.

Contents

• End-to-end workflow of AI-assisted software development at the current state of the art: from idea generation through design, implementation, and testing to deployment
• Methods for conception and idea generation with AI support, developed in full-day workshops
• Capabilities, limitations, and inner workings of large language models as a basis for using them in development
• Architecture of a coding harness: how an assistant system organises tools, context, and test loops, and how it can be adapted to one's own requirements
• Comparison of execution environments: locally hosted models, scientific cloud infrastructure, and commercial providers
• Work on a deliberately demanding project task, from finding a solution to cloud deployment
• Documentation, presentation, and critical reflection of the team's own AI-supported development process

Competencies

• carry out a software project from initial idea to deployment with AI support,
• apply AI tools purposefully in conception, design, implementation, and quality assurance,
• explain how large language models work and derive their possibilities and limitations from this,
• operate, assess, and adapt a coding harness to their own requirements,
• compare locally hosted and cloud-based models in terms of capability, cost, data protection, and practical viability,
• present interim results and outcomes to an audience and justify their design decisions,
• document and critically assess their own use of AI in the development process.

Literature

• M. Kofler, B. Öggl, and S. Springer, AI-Assisted Coding: The Practical Guide for Software Development. Bonn, Germany: Rheinwerk Verlag, 2025.
• I. Sommerville, Modernes Software-Engineering: Entwurf und Entwicklung von Softwareprodukten. Hallbergmoos, Germany: Pearson Deutschland, 2020.
• D. Jurafsky and J. H. Martin, Speech and Language Processing, 3rd ed. draft, 2025. [Online]. Available: https://web.stanford.edu/~jurafsky/slp3/
• C. Huyen, AI Engineering: Building Applications with Foundation Models. Sebastopol, CA, USA: O'Reilly Media, 2025.

Lecturer

• Prof. Dr. Sebastian Steger
• Adrian Lehmann
• Prof. Dr. Kai Eckert
• Prof. Dr. Anuja Hariharan

Recommended Previous Knowledge 

• Solid programming skills in at least one language, ideally Python.
• Software Engineering 1 and 2, or comparable knowledge of the software development process.
• Confident use of Git and the command line.
• Willingness to get familiar with a cloud environment and with running models locally.
• Willingness to work independently and continuously on the project throughout the semester.

Module Details

Weekly Hours (SWS) 

Work Load