Worked in the CloudIQ Analytics team alongside data scientists building ML models for capacity and performance forecasting and service requests' predictions for customer storage systems.

Developed a Math tutoring tool - MathSpring - as a member of a research team at the Center for Knowledge Communication at UMass Amherst.

Improved the Administration tool used by teachers and content developers to create Math problems and tutoring strategies for a class of students.

Designed the architecture for a report generation tool which enables easy creation of reports from existing data in any desired format for analysis by the management team to help drive decisions on new customer acquisitions, customer support, developing marketing & sales strategies and monitor changes in status of existing customers.

Implemented the architecture end-to-end by writing highly flexible Python modules for handling large amounts of data in the backend with support for historical analysis and creating reports in various formats including tables, charts, summaries and corresponding Angular modules on the UI to view the reports.

Did a PoC on performance impacts of using JSON in Postgres vs using MongoDB which has native support for JSON. Resulted in a performance improvement of 20x using MongoDB.