James Flemings

Hello! I am a fifth-year CS PhD student at the University of Southern California advised by Murali Annavaram in the SCIP (Super Computing In Pocket) lab. I’m graciously funded by the NSF Graduate Research Fellowship. My research addresses privacy concerns in Large Language Models (LLMs). My research aims to make Large Language Models (LLMs) safe to train and deploy on sensitive data. I work along three complementary directions: (1) developing formal, provable guarantees for the data used to train and query LLMs, (2) building LLM agents that make contextually appropriate data-sharing decisions, verifiably aligned with users’ privacy preferences, (3) adaptive policy generation to guarantee agent actions comply with privacy, security, and safety properties.

Previously, I interned at Google in the Federated Learning and Analytics team during Summer 2025, investigating personalized privacy in agents. I also interned at TikTok during Summer 2024 in the Privacy Innovation Lab, exploring hallucination and privacy in langauge models. Before graduate school, I received my BS in computer science and mathematics at the University of Alaska Anchorage in 2022. While there, I pursued research in a wide range of areas, including monochromatic colorings, charged coupled devices, federated learning, and out of distribution data performance in language models.