Wow, this is good. I gave it "Configure your pipeline state object with a compute shader that consumes uniforms and vertex buffers." and it came up with this:

Note that the compiler will add a compute shader at compile time and require you to configure your pipeline state every time. This is the equivalent on a more traditional GPU where shader compilation occurs between the GPU and CPU. For example if you want to set the shader with a single pass on a CPU it is reasonable to compile it each time after GPU rendering, and even compile it once on the GPU. This is an extension of the parallel compile-to-script: run-to-script approach.

It's a very compelling illusion of actually understanding the topic. I've seen less coherent responses on Stack Overflow and, dare I say it, Hacker News.