Switch Jockey is an application made for the web that uses machine listening to analyze real-time audio and determine high level features that are then used to conduct the switching of audio-reactive shaders written with the OpenGL Shading Language. The purpose of the project was to create a system that responds intelligently to different types of music, either produced by a live coding artist or by playback of a recording, in order to decide on the most appropriate visuals from a set of options. It is an attempt at increasing the accessibility of interesting live visuals to help give audio artists the ability to relatively easily include these in their performances. It is also an attempt to build a new pathway for collaboration between audio artists and visual artists. To just create visuals for music and have an intelligent system control the timing of performance playback reduces the need for anyone to be physically present to control the video switching.