The inspiration for this project happened when we were going through CapitalOne's ATM API, and realized that there was no way for an employee or user to intuitively tell where these ATM's were, how much they had, and other pertinent data. We decided CapitalOne desperately needed a visual representation of their ATMs, especially ATM's that are running low on cash.

How it works

It works by processing the geocoded ATM locations through a ruby processing script, which passes the data to the WebGL globe, which then renders the points across the nation. The WebGL globe is supported by a python webserver.

Challenges I ran into

The CapitalOne API UI on the website was broken, so we were unable to start until 1AM this morning! :P

Accomplishments that I'm proud of

Building an insanely cool visual representation of otherwise very difficult to understand data.

What I learned

I learned a lot about JavaScript and Ruby, particularly useful methods and techniques for fast data processing.

What's next for CapitalOne ATM Visualizer

Up next is adding extra functionality and connections within the globe to enable greater utility from the visualization - having the globe smartly display which ATM's are running out of cash, show other competitive bank ATM locations, and enable support for flow of funds between ATM's or branches.

Share this project: