During my internship at the New York Times, I built a tool to generate videos and GIFs automatically from New York Times articles. The videos are intelligently generated, but we also built a fully-featured in-browser video editor to allow content producers to tweak them. We create the videos on the front end, and this makes the application as a whole very scalable.
Using D3 and NYPD-issued stop-and-frisk data from the last 13 years, we built a classifier to display the probability of being stopped relative to an average New Yorker based on different demographic data. Then, we compared this against income and demographic data in different areas.
I am one of the most active members of the hackatbrown webdev team. Our website serves over 2200 applicants.
This project (Java) handles real-world map data and provides path information. You can pan around and click points on the map, or you can use the input fields to input cross-streets and find paths between them. It connects to a traffic server and dynamically updates clients with traffic information. The project is backed by a few interesting data structures.
The project uses multiple threads to ensure that the frontend GUI remains responsive no matter what kind of requests are being made to the server.
One of my first projects, it had a pretty large feature set. Hosted on EC2, and with a java swing client app, The backend uses a lightweight H2 database for storing cards and their tags. It allows you to import cards in several ways. When you're finished, you can export the cards as an iTunes playlist or as MP3s to sync with your phone.
Crowd-controlled cookie crushing. Websockets, Streaming Video, Intel Edison.
This is a little app to help you pair off members of your team for lunch dates.