How does a computer identify pictures of cats? What about translating speech from one language to another? Or drive a car? Creating lifelike photos of people who don’t exist? These are jobs for deep learning. Using neural networks, deep learning is a specialization of machine learning with a predisposition for perceptual tasks requiring lots of data. If this sounds hard, it’s because it is! However, the Python community has pitched in and offered two libraries which are the focus of this talk. Keras is an high level conceptual API which allows developers to focus on the why of deep learning projects instead of the how. For those who need more control, or to invent something new, TensorFlow is a lower-level library which is also a backend for Keras. Between the two there is something for everyone available in the exciting new world of deep learning.