About this event
AI-assisted Data Manipulation & Programming: User Experiences, and Neuro-Symbolic Techniques
Speaker: Sumit Gulwani (sumitg@microsoft.com)
Abstract: The generative AI revolution has enabled a broad range of applications, using natural language as intent expression, and powered by numerous prompt engineering innovations and agentic architectures. In this talk, I will advocate for other forms of intent expression in addition to natural language, such as input-output examples, static and temporal context, and broken artifacts for repair rather than generation from scratch. I will also advocate for leveraging neuro-symbolic techniques, which combine the power of LLMs with logical-reasoning based symbolic techniques, to build more effective solutions for specific verticals. And of course, I will also share some prompt engineering tricks and techniques and agentic architectures, including providing an escape mechanism to reduce hallucinations, smart example and documentation selection for in-context learning, and meta-reflection to improve prompt instructions. This will be discussed in the context of applications in data manipulation (string transformations, table extraction) and programming (debugging, hint generation for learning, and NL2Code for low-resource languages).
Bio: Sumit Gulwani is a computer scientist connecting ideas, people, and research & practice. He invented the popular Flash Fill feature in Excel, which has now also found its place in middle-school computing textbooks. He leads the PROSE research and engineering team at Microsoft, driving advances in AI-assisted programming and incorporating them into various Microsoft products including Visual Studio, Excel, PowerQuery, PowerApps, PowerAutomate, Powershell, and SQL. He is a champion of storytelling trainings and initiatives within Microsoft. He initiated a novel research fellowship program in India, a remote apprenticeship model that scales up impact, nurtures globally diverse talent, and grows research leaders. He has co-authored 13 award-winning papers (including 3 test-of-time awards from ICSE and POPL) among 150+ research publications across multiple computer science areas and delivered 70+ keynotes/invited talks. He is a Fellow of the ACM and AAIA, and received the Max Planck-Humboldt medal in 2021 and the ACM SIGPLAN Robin Milner Young Researcher Award in 2014 for his pioneering contributions to AI-assisted Programming. He obtained his PhD in Computer Science from UC-Berkeley, and was awarded the ACM SIGPLAN Outstanding Doctoral Dissertation Award. He obtained his BTech in Computer Science and Engineering from IIT Kanpur, and was awarded the President’s Gold Medal, and the Distinguished Alumnus Award.
To Join the talk:
Register in advance for this webinar:
https://nsf.zoomgov.com/webinar/register/WN_W3H0zKU4TJ6vf907ErMSyQ
Or an H.323/SIP room system:
H.323: 161.199.138.10 (US West) or 161.199.136.10 (US East)
Meeting ID: 160 636 9558
Passcode: 921601
SIP: 1606369558@sip.zoomgov.com
Passcode: 921601
After registering, you will receive a confirmation email containing information about joining the webinar.