Django’s great for building web apps. Jupyter’s great for sharing reproducible analysis. Altair’s great for making analysis interactive, and easier to understand visually. Here’s how to use them all together. For a large part of last year, I was working in a small, cross disciplinary team in a business information company, building data products to help make sense of public spending data. Our shared language across the team was python, and over the year, we came up with a workflow that let our analysts and data science specialists work with Jupyter notebook to carry out analysis, and build interactive data viz using Altair to explore data, before taking the same code used to generate these charts, and incorporate them in the main django app as data products for our customers. This talk outlines why these tools are useful together, explaining how Altair packages up some of the most recent advances in thinking about data vizualisation techniques, and makes them accessible to python developers for exploratory analysis of data using Jupyter and Django together. We’ll finish by showing how, once you’re happy with them, you can then integrate these same interactive visualisations into an existing django application, and make them accessible to your users.