Completed my thesis under Prof. Radhika Mamidi on “Towards Improving Vision-Language Multimodal Systems” with publications across visual question answering, multimodal machine translation, and multimodal sentiment analysis, leveraging large language models. Additionally published research in other areas in NLP including code-mixing, table understanding, and machine translation.

Worked as a full stack engineer and built a rule editor in the revenue computation platform using Go backend and React (fusion.js) frontend. The code editor included extended features like syntax highlighting, and auto indentation.

Worked as a part of Mixxx organization and re-designed the Effects Engine Architecture written in C++ to tackle long term tech-debt and finally enabled the system to onboard pending feature requests blocked on infra limitations.