Rama Ramakrishnan teaches this session of MIT's 15.773 Hands-On Deep Learning course, recapping the standard neural network training flow before walking through building a convolutional neural network step by step in Colab. The class works through the practical mechanics of setting up a deep learning pipeline for computer vision, from data handling to model construction, with an emphasis on hands-on coding rather than abstract theory. Ramakrishnan narrates each step as it appears on screen, making the session useful as a followed-along coding session as much as a lecture. Part of a Spring 2024 MIT OpenCourseWare series aimed at giving students working familiarity with deep learning tools through direct implementation rather than lecture-only instruction.