Since the market entry of GitHub Copilot (2023), generative AI (GenAI) tools have rapidly become part of everyday software development. What began primarily as code completion has evolved into broader support through chat-based interaction, code explanation, refactoring suggestions, test generation, and more. Early research indicates that developers do not use these tools in a uniform way—usage strategies, interaction patterns, and perceived benefits vary substantially depending on the person, task type, team setting, and development environment.

This thesis is positioned as a topic area rather than a fixed topic: it investigates effective use of GenAI tools in software development. The exact focus will be defined later, together with the supervisor, depending on the student’s interests, background, and skills.

The thesis may explore one or more of the following directions (non-exhaustive):

• Usage strategies and interaction patterns: How developers prompt, iterate, verify, and integrate suggestions into their workflow
• Task-context dependency: Differences in effective use across tasks (e.g., debugging, refactoring, documentation, testing, learning unfamiliar codebases)
• Quality assurance and risk handling: Validation practices, trust calibration, dealing with hallucinations, security/privacy concerns
• Tool ecosystem comparison: Copilot (autocomplete + chat) and other GenAI tools used in practice; reasons for choosing certain tools or setups
• Perceived productivity vs. actual effectiveness: What developers consider “effective,” how they assess outcomes, and what trade-offs they accept
• Emerging best practices: Informal guidelines, heuristics, do’s/don’ts, and patterns that can be translated into actionable recommendations

To develop a qualitative, practice-grounded understanding of what “effective use” of GenAI tools looks like in real software development work, and which strategies, conditions, and constraints shape this effectiveness. Depending on the agreed focus, the outcome may include a taxonomy of usage strategies, a set of developer archetypes, or evidence-based recommendations for practitioners and/or organizations.

The thesis should follow a qualitative research design, centered on:

• Semi-structured interviews with software developers (e.g., across different experience levels, domains, and team settings)
• Qualitative analysis (e.g., thematic analysis, coding, or grounded-theory-inspired approach—exact method to be agreed)

Candidate profile

• Interest in developer tooling, human–AI interaction, and empirical research
• Comfortable communicating with practitioners and conducting interviews in English (or local language, depending on context)
• Basic familiarity with qualitative methods is beneficial
• Beneficial: Background in software engineering / computer science / information systems (or related)

Notes

This is intentionally framed as a topic area. The final research question(s), scope, and target population will be defined jointly during the initial phase of the thesis.

If you want, I can also provide a slightly more “official” version with placeholders (institution, chair, supervision, duration, start date, contact), or a shorter version for posting on a website/Slack.

Relevant literature:

Burton-Jones, A., & Gallivan, M. J. (2007). Toward a Deeper Understanding of System Usage in Organizations: A Multilevel Perspective. In Source: MIS Quarterly (Vol. 31, Issue 4). [http://www.jstor.orgStableURL:http:/www.jstor.org/stable/25148815Accessed:30-03-201506:39UTC]http://www.jstor.orgStableURL:http://www.jstor.org/stable/25148815Accessed:30-03-201506:39UTC

Burton-Jones, A., & Grange, C. (2013). From use to effective use: A representation theory perspective. Information Systems Research, 24(3), 632–658. https://doi.org/10.1287/isre.1120.0444

Mueller, L., & Bruhin, O. (2025). Developer Resistance to Generative AI Adoption: Identifying Barriers in Software Development. ICIS 2025 Proceedings. https://doi.org/https://aisel.aisnet.org/icis2025/impl_adopt/impl_adopt/3

Trieu, V. H., Burton-Jones, A., Green, P., & Cockcroft, S. (2022). APPLYING AND EXTENDING THE THEORY OF EFFECTIVE USE IN A BUSINESS INTELLIGENCE CONTEXT. MIS Quarterly: Management Information Systems, 46(1), 645–678. https://doi.org/10.25300/MISQ/2022/14880

If this thesis topic description sounds interesting to you, please reach out to the topic owner Leon Müller (leon.mueller@unisg.ch).