Large Language Models (LLMs) are revolutionizing software development, offering capabilities from code generation to complex problem-solving. In this new world, where repositories are increasingly filled with data generated by AI and used to train the next generation of models, what is the role of Mining Software Repositories (MSR) research? This talk will explore MSR research through questions relevant before, during, and after interactions with LLMs. For example, on the input side, MSR can assist in ensuring the diversity and quality of the data that trains these models, as well as develop methodologies to detect and correct biases within these datasets. As developers interact with LLMs, MSR can examine the dynamics between developers and LLMs, uncovering best practices that facilitate effective communication and collaboration with these AI systems. On the output side, MSR researchers can strive to improve the output and identify areas where AI surpasses human efficiency or introduces innovative solutions, while also pinpointing where the nuanced understanding and creativity of human developers are indispensable. However, we cannot lose sight of the broader picture: the integration of LLMs into software development raises fundamental questions about the future of our field. How does this paradigm shift affect software development processes, the experience of developers, and the focus of educational and research efforts in software engineering? MSR research is essential in addressing these questions, offering a data-driven GPS for navigating the complex landscape of AI-driven development.

Christoph Treude is an Associate Professor of Computer Science at Singapore Management University. His primary research goal is to improve software quality and the efficiency of software developers, particularly by improving access to relevant information.

Treude’s research methodology combines empirical studies with the development of tools and approaches that consider the diverse natural language artifacts in software repositories. He has authored over 150 scientific articles with more than 250 co-authors. His work has received recognition, including an ARC Discovery Early Career Research Award (2018-2020) and funding from industry leaders such as Google, Facebook, and DST.

Treude has received four best paper awards, including two ACM SIGSOFT Distinguished Paper Awards. Before joining Singapore Management University, he held senior lecturer positions at the University of Melbourne and the University of Adelaide. He also worked as a postdoctoral researcher at McGill University, the University of São Paulo, and the Federal University of Rio Grande do Norte.

Currently, Treude serves on the Editorial Boards of the IEEE Transactions on Software Engineering and the Springer journal on Empirical Software Engineering. He also holds the role of Open Science Editor for the Elsevier Journal of Systems and Software. He has chaired conferences such as ICSME 2020, ICPC 2023, and TechDebt 2023 and regularly participates in software engineering conference program committees.