About me
Hi everyone!
I am currently pursuing my MSCS in the Machine Learning Track from Columbia University. During this time, I have had the opportunity to deep dive and explore deep learning systems for vision and language, large language models and high performance computing and performance optimization. Currently, I am interning at Synopsys where I am developing multivariate transformer models to solve gate sizing problems in a circuit path for chip designs.
Prior to my Masters, I was a Software developer in the International Machine Learning team at Amazon where I helped build the search ranking models and ML engineering stacks for the data discovery project. Alongside, I contributed several features to my team’s inhouse ML platform, especially related to model monitoring.
I am very curious about robustness and explainability in self-supervised models. To this end, I am researching on finding backdoors and vulnerabilities in models like CLIP and GPT-3 with Prof Junfeng Yang and Yun-Yun Tsai using reasonable prompts. I am highly interested in ML safety and aligning model training with human qualities and reasoning.
I graduated with a B.E. (Hons.) in Computer Science and a MSc. (Hons.) in Economics from BITS Pilani, Goa, India in 2020. I have had the good fortune of completing my Bachelor’s thesis in Maastricht University, Netherlands, where I was advised by Dr Amrapali Zaveri and Dr Deniz Iren. Here I worked on intelligent crowdsourcing methodology to effectively detect lung cancer tumors using active learning.