Mitigating risk in quality is a complex challenge for every test engineer and even more so for those in the hardware ecosystem due to the ever-growing support matrix. Join Martina Sourada as she talks about how Deep Learning models have the potential to ameliorate the issue of scale and risk. Martina will provide some examples of better efficiencies already being put into a hardware test production pipeline, as well as provide an overview of how Deep Learning will help with some of the more complex challenges we all face going forward. A useful test toolbox is not comprised of one tool alone. Automation has long been a push in all software spaces, but for the hardware ecosystem, specific characteristics make it challenging, requiring a higher percentage of manual resources and effort. During her session, you will learn the scope and challenges of automation in graphics hardware, how she has applied Deep Learning to address some of those challenges., and based on the success of her model, how she is looking to adapt more Deep Learning models across her production pipeline.