Built an AI system using the Retrieval Augmented Generation (RAG) framework to extract Business Goals and Plans from any Business Document. Functioning as a component of the semantic parser, the system uses multiple Large Language Models (LLMs) to extract high quality goals and procedures and has been tested on documents from various domains like Banking, Insurance and Legal.

Working with Professor Ponnurangam Kumaraguru and a group of researchers who study, analyze, and build different aspects of social systems (e.g. social web systems like Twitter, Facebook), including their security and privacy.

Added multiple new features to Qualcomm’s interactive Device Tree tool. Added Device Tree Overlay support for Device Trees used during Boot, enabling the addition of new nodes to the tree and modifying properties in the existing tree. Extensively prototyped and tested Device Tree overlays across various chipsets in the Snapdragon family

Built ‘Reddit for Research’, a web application where users can read and collaborate on research in a seamless manner. Developed a Machine Learning based search bar. Also involved in the full stack development of the product comprising of a document reader, integration of an annotation and commenting tool and the ability for multiple users to comment and discuss. Was also involved in the deployment of machine learning workflows on kubernetes using kubeflow.

Built a Content based Recommender System using Deep Learning and NLP for Qualcomm's bug and feature tracking tool (Orbit) which uses both bug/feature description and user history to provide accurate suggestions for customers and users.Developed a hierarchical clustering algorithm for the same tool, providing relevant suggestions based on CR(Change Request) Title, reducing CR creation time and redundancy.Finalist of IdeaQuest 2019, Qualcomm's in-house, innovation driven contest.