I am a first-year graduate student pursuing my Masters in Computer Science at New York University, NY. My interests revolve around full-stack software development, cloud computing, Machine learning and Data Science. Currently, I am working as a Reseach Assistant at NYU Steinhardt School of Culture, Education, and Human Development under Prof. Peter Halpin and Prof. Yoav Bergner to create a tool which helps bringing real-time collaboration into online learning based on the Openedx platform and Mozilla’s togetherJS library. I am also working on a few Machine learning projects as part of my academia, and working on deploying the models on IBM Cloud and AWS.
I completed my Bachelors in Computer Engineering from St. Francis Institute of Technology, Mumbai in 2015. During my undergraduation, I worked under Dr. Vikram Shette on developing a voice activated, dynamic braille script generator which served as a self-tutoring braille device for the visually impaired students. The project was regarded as one of the best projects in 2015 in an MIT-Sponsored National Level competition, TechTop India, 2015.
I have 2 years of professional experience as a software developer. On an individual level though, I have been developing software applications and side projects for a long time(6+ years).
I would describe myself as a confident, passionate software developer who genuinely loves programming, developing applications and everything about the computer science domain. One of the things I pride myself for is that I am an extremely efficient learner, who isn't afraid to fail if it means trying something challenging! Since the last few years, I make it a point to study something new(be it a programming language, a tool or a hobby!), every few months(or weeks).
I am currently working on statistical analysis on tweets generated during the demonetization event(it kinda was an event right?) in India, and perform sentiment analysis and see how people reacted before, during and after the decisions were made. I am using MongoDB on the backend, and working on IBM Cloud provided in my Big Data Course.Machine Learning Data Science R MongoDB
A blog application using MongoDB, ExpressJS, NodeJS, PassportJS and Bootstrap. Features include: adding blogs, creating blogs, user authentication using PassportJS, Adding images in blogs, syncing all the blogs with mongodb, Can view other users blogs but cannot edit delete it!NodeJS ExpressJS PassportJS MongoDB Bootstrap Code repo
I always try and automate tasks at my work place. One such problem during my office hours at my on-campus part-time job required me to count things of different categories. Since I couldn't find a proper tool and since I was too lazy to do it manually, I decided to make a solution using my skills.
Solution: An android app to count things, different categories simultaneously!
Since I was on it, as soon as I was done, I decided to make a similar app using Angular 2, just to warm my new learned skills in Anuglar and TypeScript.Android Angular 2 More info Android Angular 2
Built a Machine Learning classifier which analyzed twitter accounts and classified them as bot or human. Used a soft voting classifier on top of Gradient Boost, Random Forest and Adaboost Classifiers. Used Tableu and Python's Matplotlib library to generate graphs and aid analyzing patterns in the data. Used Python’s numpy and pandas libraries to clean and standardize data. Achieved 98.5% accuracy on the test dataset.Machine Learning Kaggle Scikit Learn - Python Twitter API More info Code Repo
Built an AI player with 3 difficulty levels: Easy, Medium, Difficult. Reduced AI response time to 10 seconds(maximum). Used Pythons Pygame library to build the system. Developed an interactive user-interface.MiniMax Algorithm PyGame Python Code Repo
Some Projects I did in Python as part of Programming with Python Udacity course:
1. Prank Secret Message decoder using bulk file rename
2. Movie website with titles and in-page embedded trailers
3. Profanity checker in documents
4. Take a break reminder
Developed an entire GRE Test series for a client from scratch. Adaptive Nature of the test allowed adjusting the difficulty of the upcoming sections of the test based on the students' performance in the current section.
Designed and developed a voice activated, dynamic Braille Script generator which served as a self-learning tool for the visually impaired, while working with people from Industrial and Mechanical domain. Developed an android application for audio input, to be then sent wirelessly to an Arduino processor, which sent signals to the appropriate pins on the device to actuate or retract to display the input character. Worked with a group of visually impaired students to test the productTop 20 projects in India, 2015 Android Shape Memory Alloys Braille Tutor Arduino prcessors More info Video
Implemented a database management system for efficient and safe transactions between customers and clients. Used technologies: HTML5, CSS, Java Database ConnectivityHTML5 CSS3 Java Database Connectivity Java PHP MySQL
Created a web application using PHP where users upload, download, view and share all types of files online.HTML5 CSS3 MySQL Java PHP bootstrap jQuery Code repo