Streaming platforms are massively complex software systems, involving numerous technical and behavioural variables that are very difficult to predict and/or outside of our control, the art for all streamers is hiding that complexity from viewers. Part of that art is how these systems are operated and managed day-to-day and prepared for major peaks in consumption demand, while generally being resilient to failure. This session will explore how we operate ITVX, plan for major events and prepare our playbooks to handle failure. We’ll cover some of our preparatory work, the technical resiliency patterns used and some of the wider practices we employ during a major event.