Varshini Elangovan
Graduate Student in Machine Learning, CMU
Hi! I’m Varshini Elangovan, a Master’s student in Machine Learning at Carnegie Mellon University, graduating in December 2026. My research interests include LLM post-training and alignment, AI safety, multimodal reasoning across VLMs and VLAs, and reinforcement learning.
At CMU’s Forge Lab, I work with Prof. Virginia Smith and Aashiq Muhamed on post-training interventions for LLM alignment, with a focus on understanding and suppressing unsafe persona-driven behavior using consistency training.
At the Human Sensing Lab, I work with Prof. Fernando de la Torre on multimodal models: efficient 3D Visual Question Answering with 2D VLMs in collaboration with Meta Reality Labs and VLAs in collaboration with Fujitsu Research.
Previously, I was at Google Hyderabad for 3 years as a part of the Global Payments Platform Risk team, working on fraud mitigation for peer-to-peer payment transactions in Google Pay and Google Wallet.
I’m currently seeking full-time roles in ML research and engineering starting January 2027.