- This event has passed.
Kahlert School of Computing Distinguished Lecture Series
October 17, 2024 @ 3:30 pm - 5:00 pm
Microsoft Research
October 17, 2024
3:00 refreshments
3:30 lecture
Host: Ganesh Gopalakrishnan
Safe Natural Language Programming
Abstract: AI agents have recently demonstrated impressive human-level capabilities in various software engineering tasks. More impressively, these capabilities are increasing at an unimaginable pace, with qualitative step improvements every few months. What does this mean for the future of programming? Is English indeed the next programming language? Do we still need programing language research? At Microsoft Research, some of us have long predicted these AI advances and have been working on answering these questions. In our view, contrary to what some might believe, this is the time for researchers to double down and build the foundations that will shape the future of programming.
We believe that the world is inching towards safe natural language programming. Just as type-safe programming shields programmers from the complexities of low-level programming, safe natural language programming will shield future programmers from the complexities of high-level programming. We foresee a future where humans express their intent interactively and naturally to generate a precise specification, which is converted through a combination of symbolic and AI tools to a program that implements the specification provably correctly and performantly. Humans can test, debug, performance engineer, and maintain programs through natural interaction without looking at code, just as type-safe programmers perform these tasks today without looking at assembly.
I will describe ongoing research projects at MSR that builds towards this vision and open research problems that remain.
Bio: Madan Musuvathi is a Partner Research Manager at Microsoft Research leading the RiSE group that focuses on research in programming languages, formal methods, software engineering, and high-performance computing. His research has produced several software reliability and performance-engineering tools that are widely used within Microsoft and other companies. He received the CAV award in 2023 for his fundamental contributions to the field of computer-aided verification. He has won distinguished paper awards at several conferences including PPoPP ’21, SOSP ’19, and OSDI ‘04. One of his co-advisees won the 2012 ACM SIGPLAN Outstanding Doctoral Dissertation Award. He co-chaired the Program Committee of ASPLOS ’24. He received his Ph.D. from Stanford University.