I am a master's student in the Machine Learning Department at Carnegie Mellon University, working on flow- and diffusion-based generative models across language and control. I am currently advised by Prof. Nick Boffi on continuous flow-based language models that enable efficient few-step and one-step text generation, and by Prof. Max Simchowitz on sample-efficient reinforcement learning for full-finetuning flow- and diffusion-based policies in robotic manipulation. My research focuses on how flow-based generative models can be made fast, stable, and expressive — with the broader goal of establishing flows as a unified modeling framework for structured generation and decision-making.
Before CMU, I studied Computer Science at IIT Bombay. Outside of research, I'm usually buried in a book (here's my reading list), chasing down a new film, or wandering Pittsburgh in search of good food. I also write on Medium now and then, mostly about ideas I can't quite shake.