I have a mart that times out on my laptop. I was able to speed things up somewhat by reconfiguring joins and whatnot. I'm looking for advice to speed it up further. One idea I had was to convert down stream views to tables so they're formed during DBT rather than on the fly. Is that a standard thing or am I going against the grain?

Did you mean this for a different slack instance? I have stg, int and marts, typically my stg and int models are views but I can swap some of those ints to tables if this speeds things up. If swapping your stg from views to models makes a big difference then possibly your staging models are too complex?