Walking Manhattan
This project is about walking as many of the streets of New York as possible and keeping track of my progress. I started this project after moving to New York in 2017 and wanted to experience as much of the city as possible. The data for this project is collected via a free GPS recorder (GPX Tracker) for my phone whenever I am taking a walk or walking back from the office. The GPX data is first cleaned using an editor (GPX Track Editor) to make the rendering less intensive and then converted into less verbose format that also strips out any timing information for privacy reasons using a Python script. The file is then parsed using JavaScript and rendered using the Google Maps API at the same time.Python, JavaScript
Flight Map
This is another project of me learning how to use Javascript and the Google Maps API, plotting all the flights that I have taken in my life. The data is using the Google Maps API offline to retrieve longitude, latitude pairs based on the international airport codes and processing these into a data file. That file is parsed in Javascript and rendered using the Google Maps API.Python, Javascript
1 second everyday

The script I created crawls a Dropbox directory and interfaces with ffmpeg to extract one second for each of the videos in the directory. There is configuration file to determine which second of a longer video to use and whether to apply other transformations (mostly converting portrait into landscape, rotating an image by 180 degrees, and upscaling older videos). It then adds the date of that video (based off the file name) in either white or black text and reencodes all one second videos into one long video. All of the original videos are kept to prevent a quality degredation that would otherwise occur if new frames are added to an existing file. I have been doing this since December 2013 and unfortunately missed many days since, but it is an experience that I can recommend to everyone.
GitHub, Python
AAA Combat Simulator

GitHub, C++ with Qt
Combat Ally

GitHub, C++ with Qt
APOD Wallpaper

GitHub, Python
Stammbaum
This application is a simple graphing tool for a family tree. It was meant as a way to easily collect information about distant family members an keep all information organized in a custom database format.GitHub, C++ with Qt
Character Status
This application was an iOS application on the AppStore used to replace a physical character sheet in Dungeons & Dragons 4th Edition. It kept track of stats, skills, afflictions, hitpoints, armor class, powers, and all other important aspects of a character. It has since fallen into disrepair and has been removed from sale in the App Store.Objective-C