I am interested in designing DL methods for Speech/CV/NLP domains as well as areas like Recommender Systems that combine multiple domains. I like the challenge of learning useful representations for new data and counter the problem of less labelled data by using techniques to extract as much useful information from unlabeled data. To this end, I am interested in exploring data augmentation, transfer learning, few-shot learning and self-supervised learning methods. With the boom in ML/DL/AI, I feel it is important to know how to apply these techniques for different problem domains. I am looking forward to learn more in this front.

Currently, I am working at Meta as a Machine Learning Engineer. Prior to this, I was working with the Bixby team on developing DL models for trigger word detection.

I was a graduate research student at the Autonomous Insect Robotics lab (AIR lab), University of Washington , Seattle. My work was on developing Computationally constrained vision based sensing for insect scale robots.

A detailed description of my research internships, projects and online courses can be found in my Resume and the links provided below.