cv

Data Scientist

Data Science for Common Good

Worked with UMASS CICS Center for Data Science(CDS) and Springfield Public School(SPS) system on the development of a tool to identify and improve course recommendations to students based on a model of success created from course trajectories of previous students. We used a Bayesian network to model the causal relationship between student course trajectories and their success probability.

• Bayes Networks
• Computing for Common Good

Research Intern

Google Summer of Code with C3G Canda

The problem of inferring unobserved values in a partially observed trajectory from a stochastic process can be considered as a structured prediction problem which is traditionally conducted using heuristic-based Monte Carlo methods. In this project, I studied the application and effectiveness of learning proposals for sequential importance samplers using the concept of reinforced variational inference. Work done was published in ICLR 2019, Deep Reinforcement Learning Meets Structured Prediction workshop. The final product is an open source framework that provides a convenient interface for researchers to create and manage Reinforced Variational Inference experiments.

• Reinforcement Learning
• Genetics

Research Assistant

University of Massachusetts, Amherst

Currently I work as a research assistant in Bio-inspired Neurodynamics Laboratory in UMASS Amherst advised by Dr. Hava T. Siegelmann.

• Memory Models
• Mechanistic Interpretability

Intern

Flipkart

Engineered improvements to the efficiency of the company’s existing fraud detection technology using Decision Trees. Developed a skill tracker application for employees in the company. Developed a chrome plugin to enable new and existing employees to discover and learn acronyms used in the company.

• Chrome Plugin
• Web Technology
• Fraud Detection

Intern

IBM

Worked on creating a model to automatically assign teams to customer support tickets based on the description of the issue in the ticket using Natural Language Processing techniques. The project was part of the Remote Mentorship Program to create an EDI Spec Creator Application led by Mr. Manjit.S.Sodhi, IT Architect, IBM..

• Web Technology

MS/PhD

University of Massachusetts, Amherst

Neural Networks

B. Tech.

National Institute of Technology, Trichy

Computer Science and Engineering