Jupyter notebooks are data science’s favorite new tools, as they combine two things that we typically think of as separate: A human-readable part made of text, images, and even video, all of which combine to form the introduction, write-up, and summary that you’d expect to find in a paper or report, and An executable part made of code, which provides interactivity, performs calculations and analysis, and can generate live visualizations. It allows readers to explore the data, ask “what if?” questions, and see the results. Jupyter notebooks are tools for people who work with data to collaborate. IBM uses them to “distill data into insights”, Netflix uses them to scale up their ability to work with their copious amounts of data, and you can use them to craft reports that completely show your work — the data you collected, the code you use to process it, along with your writeup of your findings and reasoning. In this talk, I’ll walk you through the process of creating your first notebook, and using it to write your first data analysis report, complete with an executable part that performs the analysis, and a human-readable part that contains your conclusions. If you want to get into data science, Jupyter notebooks will be a key part of your toolkit, and this talk is your opportunity to get started!