Draw Platformer

Grand Prize Winner at HackUVA, March 2017

DrawPlatformer allows you to make your own Platformer style video game. With just a few markers and some printer paper, you can draw your own custom level. Then, just take a picture of your masterpiece, upload it to our server, and you have a playable game in under 30 seconds. We built this at HackUVA, UVA's premier 24-hour hackathon. Out of 35 competing teams, we were awarded with the Grand Prize (1st place) along with $1,000.

I made the front-end for the game browser, and used JavaScript to create the game’s functionalities, and my teammate used OpenCV + Skimage + Python to do the image processing on the backend.

Source Code on GitHub | Article by UVA's school newspaper | Original Hackathon Entry

Close Project

Raptionary

Finalist Project at VTHacks, February 2017

Raptionary is a mobile application that gets you the lyrics of any song and provides an Urban Dictionary lookup tool that can give users definitions to any slang words or phrases used in a song. The program uses the jsoup library and provides lyrics and definitions by web crawling azlyrics.com and urbandictionary.com. The app has reached over 600 downloads. Download Raptionary for free on Google Play!

Close Project

ESnap

VTHacks III, February 2016

ESnap is a mobile application that can provide a summary of an article or book page just from a screenshot or picture of a page. The program converts the image into text format using an Optical Character Recognition (OCR) API, and summarizes the text using a Natural Language Processing (NLP) API. The app has reached over 250 downloads. Download ESnap for free on Google Play!

Close Project

Trump Bump

May - July 2016

In this satirical grid-based game, play as Trump himself and dodge left and right to make sure he avoids his opponents at all costs! The longer Trump lasts, the more delegates and electoral votes he receives. Once he has enough, Donald J. Trump can finally become President of the United States of America and MAKE AMERICA GREAT AGAIN! The app has reached over 150 downloads. Download Trump Bump for free on Google Play!

Close Project

Enunciate.io

Submission for VTHacks, February 2019

Enunciate.io allows users to read our virtual children's books out loud to their computer and get real-time feedback on the accuracy of their word pronunciation. We used a Speech Recognition SDK from the Microsoft Cognitive Services API to convert user's speech to text. We also used the turn.js library to create our virtual books. We then built our own algorithm to assess where the user's speech matches the text in the books they may be reading.

Close Project

OcuTune

Best Networking App & Best Mobile Hack at VTHacks, February 2018

OcuTune is an Android mobile application that uses computer vision to analyze a picture of the user’s environment and provide an original playlist of music based on the user’s surroundings. We used the Microsoft Cognitive Services API to produce image tags of objects in the picture, and used the Spotify API to find relevant music based on the image tags and provide the user with the playlist.

Close Project

GitCheck

Finalist at Pennapps, January 2018

GitCheck is a web service that can automatically grade your GitHub projects according to standard style guidelines and provide detailed feedback on the breakdown of your major styling mistakes. On top of that, our service gives an overall rating of your profile and each of your repositories. You can see where you specifically need to improve to bring your ratings up and become a better developer. Company recruiters can also easily discover top-tier talent and experience in candidates by searching their ratings with our service.

Close Project

OcuLearn

Best Use of Google Cloud Services & Most Interesting Use of Data at HackDuke, November 2019

OcuLearn is a platform built to detect student engagement in online classrooms using Computer Vision and Machine Learning. The application collects real-time webcam data to process through Google Cloud Vision API providing principle axes head rotation measurements. These are run against our trained ML model that uses K-nearest neighbors algorithm and cluster analysis to reasonably assess the student's attentiveness.

Close Project

StoryLook

2nd Place Overall & People's Choice Award, November 2019

StoryLook collects audio as the user is reciting a story, and provides visuals pertaining to detected "entities" from the story, all in real time! This application utilized Google Cloud APIs such as Speech-To-Text to convert story audio into a searchable format, Natural Language Processing to detect and label entities mentioned in the story, and even Translate to support multiple languages!

Close Project

MassApply

13,000+ Users & $30k prize (2nd place) at Schulze Entrepreneurship Challenge
March 2020 - Present

MassApply helps students easily apply to companies, contact verified recruiters, and track their application status. During COVID-19, MassApply is helping students land job opportunities despite hiring freezes and internship cancellations. The dashboard was built using Node, React, Heroku, and Postgres. Many features have been automated, such as web-scraping direct application links and collecting company info using CrunchBase API.

MassApply Website | Product Hunt Listing | Featured Article

Close Project

PostGen

12 Monthly Paying Enterprise Customers
Jan 2022 - May 2022

PostGen is a tool that gives you verified emails of people who liked/commented on any LinkedIn post that you specify. I deployed a pipeline that sifts through a combination of web scraping APIs and lead generation tools to deliver a spreadsheet of all relevant contact information.

Close Project

OneShot Vision

12 Monthly Paying Enterprise Customers
Jan 2022 - May 2022

We wanted to hack a way for us to instantly learn about the people we're meeting. OneShot Vision uses our own one-shot face detection machine-learning algorithm, and our custom-built raspberry pi glasses to recognize and match a face to our database in real-time.

We built a headset that consists of a raspberry pi, a pair of glasses, an OLED screen, and a camera to process a feed of live image data. Using hard cascade image detection we’re able to isolate the face, and send it into our own ML model for one-shot facial detection. We utilized siamese neural networks to make the one-shot face detection algorithm. Once a contact’s face is recognized, we query our DB for the contact’s information and return it back to our raspberry pi glasses, which gets displayed on the wearer’s small non-obtrusive OLED screen. And since there’s only limited space on the screen, we’re also sending the user a text message using Twilio’s API to provide a full profile overview of the person you’re meeting.

Close Project