In this tutorial, the best practices for building large-scale reproducible machine Learning projects will be discussed. In addition, some techniques for defining, launching and visualizing a large set of experiments will be explored. On Google Colab, starting from an empty project, I will show how to compare between 4 optimizers on two datasets and obtain visualizations that are ready to be presented in a research paper or project report. By the end of the tutorial, you will be able to build a deep learning benchmark, launch large-scale experiments, visualize them, debug failed experiments, and generate publishable results for a final machine learning report