As a Data Scientist on the Merch Algorithms team at Stitch Fix, we empower our business partners to be data-driven through dashboards and reports. In many cases, our partners simply want to look at the data under several predefined lenses. While RStudio’s Shiny provides this capability, the growing needs of our business partners led to several large and unwieldy applications. My preference for data munging in Pandas led to the creation of Pyxley. Pyxley provides a set of simple widgets in the spirit of Shiny, allowing me to focus on the data and not the UI.

I spent a week figuring out a decent way to have Python write JavaScript, wrote a set of wrappers, called it Pyxley, and released it in July of 2015. The amount of interest was overwhelming, especially as someone new to open-source. In this talk, I will focus on what I think has and hasn’t worked for Pyxley. In addition, I’ll talk about what it’s like to maintain a project for the first time.