Program analysis empowers a variety of tasks, including efficient debugging and defect detection, code comprehension, and security analysis. In this talk, I will discuss the missing steps for making the analysis results fit for practical use, providing three examples from our recent work in the areas of software quality and security. Specifically, I will discuss work on helping developers debug regression failures in large complex systems and work on helping an expert security analyst identify malicious software behaviours. With these examples, I intend to show how making the analysis results appealing to human experts opens up new, previously unexplored research directions. Finally, I will touch on future work on explainability and understanding, which will become ever more important as technology becomes more powerful.

Julia Rubin is an Associate Professor in the Department of Electrical and Computer Engineering at the University of British Columbia, Canada. She is a Canada Research Chair in Trustworthy Software and the lead of the UBC Research Excellence Cluster on Trustworthy ML. Julia’s research focuses on quality, security, and reliability of software and AI systems. Her work in these areas was recognized by numerous academic and industrial awards, including Distinguished/Best Paper Awards at major conferences, CS-Canada Outstanding Early Career Computer Science Researcher Award, IBM CAS Project of the Year Award, Killam Faculty Research Fellowship, and Humboldt Foundation Research Fellowship. More info: https://people.ece.ubc.ca/mjulia/