Jason Tselentis makes the case for designing “type first”, then outlines a process for doing just that.
His premise is good:
Today, mobile first has become second nature, but in order to ensure the design works well on small to medium to large screens, you have to think “type first.”
My own habits have changed slightly over the years, but as I don't tend to design large scale sites or applications, a robust process has not been necessary. But if I were to create one, I can't see myself using Markdown or the preview modes in various Markdown editors as a starting point. Even Typecast never felt right to me … I appreciate the concept, but every time I gave it a try, it felt clunkier than my existing tools.
Instead, I could see myself creating a small barebones boilerplate to focus on type. Then I'd use Sass variables to test the various fonts in consideration for a project.
Further to that, although I love all things type, starting a project content first makes more sense than type first. For the content should dictate your type choices.
As Butterick puts it:
Good typography is measured by how well it reinforces the meaning of the text, not by some abstract scale of merit.