I am a graduate student at MILA and DIRO at the University of Montreal working with Yoshua Bengio on deep learning for high-dimensional global optimization problems. I am also part of LambdaZero as well as the broader drug discovery initiatives at MILA, to tackle fundamental machine learning problems to accelerate scientific discovery.

Prior to joining MILA, I was part of the Turing team at Microsoft, where I worked on efficient inference for large language models as well as on developing approaches for model compression. I worked on various models shipped at scale for millions of users across various Microsoft products. I was also a Visiting Researcher at MILA with Yoshua Bengio, before joining as a student. I did my undergrad at the Department of IT at National Institute of Technology Karnataka (NITK), Surathkal.

I have been fortunate to collaborate with a number of amazing researchers in my brief experience. I spent a semester working on my undergraduate thesis with Prateek Jain and Harsha Vardhan Simhadri at Microsoft Research on efficient machine learning algorithms for inference on edge devices. Prior to that, I spent the summer of 2019 at the Machine Learning Lab at Leuphana University Luneburg working with Ulf Brefeld and Uwe Dick working on inverse reinforcement learning. I worked briefly with Jakob Foerster on extending MuZero for cooperative multi-agent settings. I am also a part of For.ai where I have been able to work with some amazing researchers.

Please feel free to reach out to me via mail, twitter, or linkedin for discussions/collaborations. Code for my projects is usually available on github.