I am a Senior Research Scientist at Meta working on AI safety, investigations, and integrity. My work involves detecting, understanding, and disrupting the malicious use of AI across high-risk harm areas including scams, fraud, cyber attacks, and CBRNE-related threats. I conduct deep-dive investigations into sophisticated threat actors leveraging generative AI, automation, and evolving attack infrastructure. I also work on technical AI safety and alignment, improving model robustness, misuse resistance, and reliable behavior under adversarial conditions. Prior to Meta, I was an Applied Scientist at Microsoft working on click fraud detection and fraud prevention for Microsoft Ads.

I received my Ph.D. in Computer Science from UC Davis, advised by Zubair Shafiq. Before that, I completed my master’s degree at UC Berkeley, where I was fortunate to work with Dawn Song and Hany Farid.

Broadly, my interests span GenAI safety, online scams and fraud, and AI-based deception. I am especially interested in technical approaches for aligning AI systems with safety policies, evaluating adversarial misuse, and building scalable defenses against harmful model behavior. I have also previously written for The Times of India and Entrepreneur.