Varun B.

I’m a PhD student in Computer Science at Duke University, advised by Prof. Cynthia Rudin. My research focuses on trustworthy and reliable AI, particularly in high-stakes decision-making contexts for noisy tabular, vision, and language data. I've built sparse, interpretable models through discrete optimization, designed faithful uncertainty quantification methods, and developed tools for understanding data distribution shifts. My work has been recognized with oral awards at ICML and IJCAI (top 1-3%), and publications in venues like JMLR and 2 NeurIPS workshops amidst others. In industry, I’ve worked on representation learning, federated learning, and evaluation of large language models. I have hands-on ML engineering experience working as an AI Scientist at JP Morgan, where I built and productionized GenAI systems, contributed to open-source, and co-invented two patents. I previously graduated in the top 5% of my cohort as an undergrad at the University of Cambridge. I really enjoy working at the intersection of scientifically rigorous AI research and real-world deployment, and am always open to research collaborations in academia and applied opportunities in industry.