Hi Grady.
I've never done anything with ARKit myself but to answer your question, yes...(*but partially).
The basis of what is needed is 52 blend shapes - with precise names. And that is what is connected to drive the facial capture - in my (limited) understanding.
I believe that there are a few tools out there that will animate your Rigify Face and turn it into the required blend shapes - but a manual approach is certainly possible.
In this course, you learned all the technical things you need to know about shape keys in order to build all the shapes you need.
*What isn't covered though are the exact shape keys you need to work with ARKit. ie - what those 52 shape keys should look like and what they should be called.
Yeah.. I'd already found a site that has the shapes and names. :D I'd just noticed that NONE of the tutorials I'd looked at just used the methods you did to create the shapes... every one of them rigged the face first :D