I was working on improving the NLU for Google Assistant, with a focus on international languages.

Worked on an ML ranker for improving relevance of Bing Autosuggest.Developed pipelines for improving relevance of entity-based suggestions.

I was involved in the universalization of Bing Autosuggest experience.Worked on improving the relevance of Bing.Explored Transfer Learning in NLP by building Question Answer ranker leveraging deep learning models. Worked on model compression techniques for Microsoft QnAMaker.Was involved in Creativity and Emotion efforts for Ruuh, Microsoft's human-like chatbot.

Worked on Emotion Detection in Textual Conversations.

Was responsible for promoting the culture of Machine Learning and Data Analytics in the institute and providing Data-driven solutions to partner firms. Worked on projects ranging from fraud detection to sentiment analysis and topic modeling.

Worked on a deep learning based ensemble for reliable detection of blood vessels in fundus color images to help doctors screen Diabetic Retinopathy patients.

Worked on an image segmentation algorithm to improve face recognition in presence of cluttered background.