 I always tell my students the way you make your project shoot to the moon is by writing it in Jupiter. Professional people with PhDs on a mission to make great discoveries use an IDE called Jupiter like they're doing a school assignment. Where is the Zen of Python? Yes, we use the notebooks for our students but not for our production code. I mean for production code we also don't use Python. Or do we? Imagine somebody accidentally writes a Skynet in a Jupiter notebook. A Jupiter. Debugging in a spider, I Python, Jupiter is a pain. A pain. A pain. Of course, we only do our research in Python. Our prediction model is then running in C. Back in the years I thought AI was scary, data science required a PhD and computer vision was magic. Now it's all just five lines of code in Python. Will companies use it for data science? All of them. Nobody likes Matlab or R or SAS. Let's not mention Julia. No, no, no. This is not for recording Intel, IBM, JPMorgan, Goldman Sachs, Spotify, they all use Python in-house to predict trends of in-house coffee usage. What can I say? Automation, innovation, revolutionary pain. We don't use Jupiter. No. Well, sometimes you gotta show how your code works. Now I sound like a PhD but don't worry, I'm only a data scientist for quantum physics operations. First import numpy, then import the rest, kerasmole ins out, and we create a Skynet. I always get confused when switching between Matlab and Python. A one starts with a zero and one is a zero.