Assistant Professor

Department: Computer Science

Dr. Shruthi S. Kubatur is Clare Boothe Luce Assistant Professor at The University of Texas at El Paso. Her research focuses on making AI agents reliable in real-world deployment. She studies how agent failures develop across multi-step trajectories, how to estimate the risk of rare but costly failures, and how to trace failures back to their root causes. This work builds on her background in rare-event simulation, machine learning, and computer vision. She received a Ph.D. in Electrical Engineering from Purdue University in 2017, where her work with Dr. Mary L. Comer developed rare-event simulation methods for Markov random fields with application to grain growth in polycrystalline materials, supported by the Air Force Office of Scientific Research. She holds an M.A.Sc. in Electrical Engineering from the University of Windsor, Canada, and a bachelor’s degree in Telecommunication Engineering from B.M.S. College of Engineering, Bengaluru, India.

Before joining UTEP, Dr. Kubatur held research and development positions in industry. She was a Senior Research Scientist at Nikon Research Corporation of America, a Senior Scientist in Imaging and Machine Learning at Berkeley Lights, Inc., a Machine Learning R&D Lead at Apple, and Head of Machine Learning and Research at Happyverse. Her publications include work on rare-event simulation in images (IS&T Electronic Imaging, 2018), rare-event simulation for Markov random fields (IEEE ICIP, 2016), and a neural network approach to online Devanagari handwritten character recognition (HPCS, 2012).

Keywords of Expertise: AI Agent Reliability, Agentic AI, Rare-Event Analysis, Causal Root Cause Analysis, Machine Learning, Computer Vision, Monte Carlo and Rare-Event Simulation.