Career Profile

Pursuing Master in Computer Science in Stony Brook University. Enthused by Artificial Neural Network, Big Data, Multimedia systems and/or General Software Development

Experiences

Software Engineering Intern

May 2018 – Aug 2018
Evernote, Redwood City

• Implemented Keyboard shortcuts for Evernote’s all new web platform for performing content formatting and various operations such as new note, adding tags and expanding/collapsing notes.
• Implemented Quick switcher, an app wide context switcher, to quickly navigate and search within notes notebooks and workspaces, and sort them according to the most recently accessed items.
• Frameworks used: ES6, Javascript, React, Redux, redux-sagas, ducks, chai-js

Software Engineer

Jan 2017 – Jul 2017
Rubique, Fintech Startup

Android Application - https://goo.gl/pk6cDv
• Conceptualized, developed and deployed mobile application for Rubique
• Implemented major projects such as search and filter module, share and earn module, document upload section for KYC, Chatbot for Rubique, Meeting module, etc.
• Implemented redesign of the app along with animations and API integration with new back-end system.
• Implemented several custom libraries, such as custom wrapper around Volley, custom seekbar, custom collapsible calendar class, custom animation classes, etc.
• Technologies used: Firebase, REST APIs, Volley, Glide, SnappyDB, etc.

Software Engineer

2014 - 2017
Samsung Research Institute, Noida

• Worked in System Apps team (Clock application) for all Android Samsung flagship devices and tablets.
• Worked on the turbo speed technology in J210F, rewrote Google AOSP Clock app to follow Samsung UX/UI instead of legacy code achieving 40% faster performance with 10% better space efficiency.
• Worked in Samsung HQ, South Korea for project development of Clock application in Tab E(T561).
• Improved audio focus and flash focus (for visually impaired) of Alarm and Timer module to better handle interaction with other system applications.

Projects

PROMPT - A Tag recommendation system for the 100M Flickr Dataset(YFCC100M) leveraging personal and social contexts and multimodal analysis on images to recommend tags. Performed user-profile clustering, asymmetric tag co-occurrence probabilities of tags, neighbor voting, and random walk to recommend top 5 tags. Published in IEEE (DOI:10.1109/ISM.2016.0109).
Design-as-embeddings and their application for improving naively generated advertisement designs. - Currently working on a project for using deep learning models to improve naively generated designs. We train a variational autoencoder on dataset of design blueprints with each category as a channel that can improve any naively generated design based on essential design aesthetics learned from the dataset.
HTML Ads Scraper - A online ads scraper project to extract ads from webpages for creating a dataset of online ads. The aim of the dataset is to collect static html to capture design cues from any online ad campaign. (Selenium web driver, PhantomJS, Python)
Satellite Image Analysis for Water Crisis - Find similar regions in world based on the trends of lost or new areas of permanent water sources using satellite water dataset analysis. This could be used to predict water quality, consumption and depletion patterns of other countries by extrapolating data from United States water bodies (USGS Dataset). (Spark, Python, LSH)
Pheno-predictor - Classify phenotypes from 0.2 million Transcriptomic Features and 1.2 million Equivalence Classes output from Salmon (a genome mapping and quantification tool) using dimensionality reduction, Artificial Neural Networks and XGBoost. (Tensorflow, Scikit-learn, Python, XGBoost)
Koupe - A Ride Sharing Android Application. Worked on a startup to develop a car pooling app using Google Maps API’s and novel ride matching algorithm, to map riders and drivers heading in same direction. (Android)

Skills & Proficiency

C++

Android

Java

Javascript

Scikit-learn and Tensorflow

OpenCV

Python

Spark