Seattle, Washington, United States
I am an Applied Scientist at Amazon in the Sponsored Products org with almost 2 years of experience. I have applied state of the art Natural Language Processing algorithms at the scale of Amazon's data. I have developed various BERT Based Large Language Models including multitask models, twin tower models while dealing with challenges posed by data scale, noise, and imbalance. I have collaborated extensively within and between teams. Before Amazon I gradusted with a PhD in the DICE track of ECE Department at the University of Texas at Austin. I was advised by Prof. Constantine Caramanis and my research is on robust estimation of high dimensional probabilistic graphical models. I had gained valuable industry experience during my Ph.D. I was a Research Intern in the Core Data Science team at Facebook during the summer of 2020. I worked on developing scalable and robust algorithms for the human labeling framework. Before that, in the summer of 2019, I interned as an AI Researcher at InterDigital, where I worked on Bayesian Neural Networks for Active Learning. Prior to starting my PhD, I completed my MS in the field of Genomic Signal Processing at Texas A&M University where I was advised by Prof. Aniruddha Datta. My thesis focused on using Bayesian techniques for the estimation of cancer tissue heterogeneity. I am currently looking for full-time positions.
Building integrity systems for Ray Ban Meta Smart Glasses.
I work on developing robust algorithms for undirected probabilistic graphical model estimation.
I developed a scalable labeling algorithm robust to human errors in the human labeling framework
Computationally efficient Active Learning using Bayesian Neural Networks.