Here are a few projects I’ve worked on. More on my GitHub, of course!

Mandoline

Open-sourced a UI component that I initially built as part of the meal delivery rescheduler in Blue Apron’s iOS app. This is the first CocoaPod I’ve ever made and I learned a ton in the process (I even gave a talk on it)! Feel free to check it out and let me know what you think!

Genre Classification and Analysis, Penn Apps (Spring 2015)

EchoNest REST API, Scikit-learn, MD-5, Flask, D3.js

Plotted songs with similar features extracted from metadata in the Million Song Dataset. Used K-means clustering to group songs by different artists. My hack won Spotify’s API prize!

This project led to my undergraduate thesis, which utilized supervised and unsupervised learning methods to determine similarity of genres for pieces with the same keys and tempos. I presented my research to over 90 students and attempted to Periscope it, but alas, the technology gods were not on my side and it was not recorded. You can find the presentation slides here, though!

Product Hunt Python Library, Penn Apps (Fall 2014)

Product Hunt REST API, Python Package Index

Worked on a Python wrapper for Product Hunt’s beta REST API. This was my first time working closely with APIs — I was lucky to have a mentor who was an API engineer help me out. It is published on PyPi and you can contribute here if you’re interested!

Copocket, Y Combinator Hacks (Summer 2014)

Pocket REST API, Flask, Memcached, MongoDB

Created a web application, Copocket, that allows Pocket users to expose their reading lists to others on the platform. My team also included a privacy feature that allows a user to tag a post as private, which would exclude an article from the publicly shared list.

Sullivan-Dragas Tweet Sentiment, Cloud Computing (Spring 2014)

Hadoop, MongoDB

Ran various Hadoop jobs analyzing ~52K tweets archived by the University of Virginia Library during the Sullivan-Dragas controversy in July 2012. My partner and I analyzed the sentiment of tweets made and retweeted by @LarrySabato, ultimately summarizing his engagement during the event.

OneDir, Advanced Software Development Techniques (Spring 2014)

Django, Watchdog, SQLite

Created a file management application that synchronizes files across Unix machines. Practiced the Agile Scrum process throughout the semester with Pivotal Tracker. You can checkout the code here if you’re interested!

GitCoin, Penn Apps (Spring 2014)

MongoDB, Express, Node.js, GitHub API, Coinbase API

Worked on a web application that matches GitHub users with repositories containing issues that they could potentially solve based on previous code commits and README’s. In return for closing issues, repository owners offer bounty rewards in Dogecoin. This was a fun project that ended up in tangled in embedded JavaScript callbacks. I would love to revisit this again in the future. Feel free to let me know if you’re interested in collaborating.

Compliments, HackDuke (Fall 2013)

Android SDK, Twilio API, Google Application Engine Datastore

Collaborated on an Android application that anonymously sends an compliment by SMS to a user’s contact by providing a random set of compliments for the user to select from. I had fun writing some Android code during this hackathon — I was certainly influenced to learn more about mobile development from it!

International Security Agency, MHacks (Fall 2013)

Twitter Firehose API, Google Maps API, Alchemy API, SQLite

Worked on a web application that maps publicly visible tweets with geolocation. Using Alchemy API, we analyzed tweets to find similar sentiments regionally. This was my first hackathon project and I since attribute my enthusiasm for side projects to the positive energy generated at MHacks. While my team had unreasonable stretch goals, the entire experience of ideating and iterating for fun influenced me to cofound Hack.UVA!