Building a reliable and robust automated driving system is really hard. In the past half decade a lot of progress has been made thanks to the advancements in machine learning and deep neural networks. One of the companies on the forefront of that effort is Nvidia. While most people think of Nvidia GPUs for graphics cards and bitcoin mining, they also have thousands of engineers that works on software. Neda Cvijetic is a senior manager on the automated driving team and she hosts the Nvidia Drive Labs video series explaining some of the core principles of how they are tackling this problem.