Aaron Bembenek

I am a computer science postdoc at The University of Melbourne supervised by Toby Murray. My current focus is on building a form of AI that is simultaneously logical and intuitive; you can read more about my high-level vision here. More generally, I work in the areas of formal methods, programming languages, and neurosymbolic AI.

I earned a PhD in computer science at Harvard University, where I was advised by Stephen Chong and was a member of the programming languages group. My PhD research focused on combining logic programming and constraint solving, with applications to program analysis and synthesis; this included designing and implementing the logic programming language Formulog, a variant of Datalog for implementing SMT-based program analysis.

I have an undergraduate degree in classics from Princeton University. I enjoy reading and spending time outdoors.

Selected Honors

• McKenzie Postdoctoral Fellowship (The University of Melbourne’s flagship postdoc fellowship, starting mid-2026)
• Annual school-wide award for “Excellence in Research (Early Career)” (School of Computing and Information Systems, The University of Melbourne, 2025)
• OOPSLA Distinguished Artifact (2024)
• Caspar Bowden Award for Outstanding Research in Privacy Enhancing Technologies (PETS 2019)