1. Explaining mispredictions of machine learning models using rule induction
2. Enabling collaborative data science development with the Ballet framework
3. Characterizing dataset downsampling for AutoML
4. Generating AutoML search spaces from weak specifications
5. Providing interactive production performance feedback in the IDE
6. Establishing explicit links between runtime traces and source code
7. Conducting an empirical analysis of the Docker container ecosystem on GitHub
8. Studying performance variation and predictability in public IaaS clouds
Received Facebook Research Award and IBM Research Award.
Research Experience
Associate Professor at TU Wien; Research Scientist at Google. Leads the Interactive Programming & Analysis Lab @ TU Wien, focusing on the role of play and exploration in software and data engineering activities.
Background
Research interests include interactive programming, machine learning for software engineering, and program synthesis. Focuses on understanding and building systems that either automate or augment human ability to deal with computational task structures and artifacts.
Miscellany
Contact: juergen.cito@tuwien.ac.at; Google Scholar, GitHub, Twitter