This document provides an overview of deep learning on mobile devices. It discusses the benefits of performing deep learning directly on mobile, including improved privacy, reliability and latency compared to cloud-based solutions. It then covers various techniques for training, quantizing, and deploying models for mobile, including Core ML, TensorFlow Lite, and Google's ML Kit. The document also addresses considerations for mobile deployment like hardware performance, energy efficiency, and supporting different device types. Overall, the document serves as a practitioner's guide to developing and optimizing deep learning solutions for mobile.