Developer of DiCE, an open-source software for explaining the decisions of AI models, which is now an integral part of Microsoft's Responsible AI Toolbox.
Research Experience
Data Scientist at Everwell Health Solutions; AI Center Fellow at Microsoft Research India; Member of the Third Space research group, Dynamics Graphics Project lab, and Human-Centred Data Science Lab.
Education
Ph.D. candidate at the Faculty of Information, University of Toronto, co-advised by Prof. Ishtiaque Ahmed and Prof. Shion Guha; M.Sc. in Artificial Intelligence from KU Leuven, Belgium; B.Tech in Instrumentation and Control Engineering from NIT Trichy, India.
Background
Research interests: supporting the safe and responsible use of AI, analyzing tensions, expectations, and technical assumptions underlying AI system development; developing software and practical tools for the responsible deployment of AI in real-world settings. Current research focuses: interpreting and framing LLM reasoning for AI safety; evaluating and improving LLM reasoning beyond logical correctness.