I am a Ph.D. student at the University of Illinois Urbana-Champaign in the Siebel School of Computing and Data Science. I am working with Prof. Gang Wang at the intersection of Security \& Privacy and Machine Learning. I am particularly interested in practical adversarial attacks on large language models (LLMs). My work has been published at ACL, USENIX Security Symposium, ICWSM, and the ICDE Workshop.

I recently completed my research-track M.S. at the University of Illinois Urbana-Champaign with Prof. Gang Wang in the area of Adversarial Machine Learning with a focus on Foundation Models. Previously, I worked on watermarking fine-tuned LLM-generated text (ACL Findings ‘21) with Prof. Zubair Shafiq (UC Davis) and Prof. Tanmoy Chakraborty (IIT Delhi).

Before my MS, I did my B.Tech. degree in Computer Science from IIIT Delhi in June 2021, where I worked in Natural Language Processing (NLP), Graphs, and Security & Privacy (S&P). I was a member of LCS2 and the Complex Systems Lab. I worked on identifying anomalies in the YouTube Blackmarket Network using graphs (ICWSM ‘22). I also led the information retrieval team for RecipeDB - a worldwide repository of recipes (ICDE Workshop, Oxford University Press).

I interned at LG AI Research in summer 2024 with the Bi-lingual LLM team. I worked on alignment algorithms for LLMs with Tolga Ergen and Honglak Lee. Prior to this, I interned at Ema Unlimited in summer 2023. I worked with the Generative AI Team on the problem of converting natural language intents to SQL Commands. I have also worked as a Natural Language Processing (NLP) Engineer at Prodigal Technologies for a year, where I focused on problems of Dialogue State Tracking (DST), Information Retrieval (IR), and optimizing search queries with a specific focus on finance-related conversations. Additionally, I interned at TCS Research and Centre de Sciences Humaines (CSH Delhi).