Hello, I am a CS Grad student at University of Massachusetts Amherst. I aspire to work in an evolving engineering environment to develop tools for industry use.
I am interested in Machine Learning and Computational Neuroscience, and have hand-on experience with Python, C, C++ and Java.
In my free time, I love to take long walks, read the works of Ishiguro and Murakami or simply admire Calvin & Hobbes. I am a huge fan of Liverpool FC, and am always up to talk about soccer!
I utilized my final semester of undergrad researching at the RISE Lab in IIT-M. Over there, I worked under the amazing guidance of Dr. Mitesh Khapra on a machine learning & NLP project that aimed to generate natural language descriptions from structured data.
I worked with Python and Tensorflow, and utilized algorithms such as beam search and multi-task learning to improve the classification accuracy.
During winter 2016 and summer 2017 I interned at an emerging, one-of-a-kind startup in Mumbai that focusses on automating medical insurance claim processes.
During my internships, I worked with Python to build intelligent models to properly classify medical reports, identify sensitive data and anonymize them and pick out OCR mistakes and rectify them.
I interned at GetEvangelized, an online platform for brands to meet micro-celebrities during spring 2016.
I was primarily tasked with developing a new website for the organization. For this task, I majorly worked on heavily modifying Drupal templates for inner pages, while using HTML, CSS and JS to design the home page.
Developed an end-to-end waste management system with two others as the capstone project. The system seamlessly integrates hardware fixed on trashcans, a web dashboard and Telegram API to provide realtime updates and visualizations about waste generation to the user as well as the municipality.
View ProjectA proof-of-concept project that use Voice Recognition to detect statistical facts and searched the web to either verify or refute the claim.
View ProjectUtilized Latent Semantic Analysis to identify similar movies based on IMDb summaries. Further used k-means to provide a colourful visualization to show how different movies are related.
View Project