Bachelor-/Master's Thesis: AI-Assistants for Research Software Development

Scientific progress in many disciplines increasingly relies on custom software development. However, researchers often lack the time or formal training in software engineering, which leads to challenges in maintaining software quality, development efficiency, and long-term sustainability. Recent advancements in AI-based assistant tools open new opportunities for supporting such researchers in their day-to-day development workflows.

This thesis investigates how existing AI-driven tools can be integrated into the research software development lifecycle to support researchers. By leveraging these tools, the goals are twofold:

  1. Improve the code quality to increase its longevity.
  2. Reduce time spent on programming-related tasks and free up more time for scientific discovery.

Examples of such tools include systems for automatically generating documentation, proposing test cases, or performing automated code reviews with targeted feedback.

Topics and Approaches

This thesis is conducted in collaboration with the Chair of Software Engineering and the Quantum Technology Group. It focuses on identifying, selecting, integrating, and evaluating AI-based assistant tools within the software workflows used in academic research.

Rather than building new tools from scratch, the thesis will focus on a subset of activities:

  • Explore the landscape of available AI assistant tools, particularly those suited for Research Software Engineering.
  • Analyze which tools best support researchers.
  • Set up selected tools within the development environments used by the Quantum Technology Group.
  • Evaluate their usefulness, sustainability, and impact on code quality and development efficiency.

It may address one or more of the following research questions:

  • What types of AI assistant tools currently exist for software engineering, especially in the context of research software?
  • Which of these tools are particularly effective for supporting researchers without formal training in software development?
  • How can these tools be integrated in a sustainable and user-friendly manner into existing workflows?
  • How can we evaluate the effectiveness of such tools, and which factors are critical to their success?

Our Offer

  • Direct impact on the software development practices of an active quantum computing research group.
  • Collaboration with researchers from both software engineering and physics.
  • Hands-on experience in evaluating cutting-edge AI-based development tools.
  • Insight into the interdisciplinary field of Research Software Engineering.
  • Possibility to tailor the thesis to your interests and strengths.

Your Profile

  • Pursuing a Bachelor’s or Master’s degree in Computer Science, Software Systems Engineering, or related fields.
  • Familiarity with modern software development practices (version control, testing, documentation).
  • Interest in AI-based developer tools and their practical applications.
  • Ability to independently evaluate tools and communicate findings.
  • Prior experience in Python may be helpful.

Contact

For more information, please send a short description of your background and an overview of your grades to Marc Schmidt schmidt@se-rwth.de.