I know a few people who've made good money immersing their hands in this pile of rich manure as consultants, so I guess it all comes down to what you individually are willing to do for some cash.
I did it for a few years after I graduated. It paid about a 10% premium over what other dev jobs in the area paid, and you never had to do anything super technically challenging, but you were pretty much at the whims of the sales org without any kind of product management in place. Plus the owner of the stack you worked with was openly antagonistic to the fact that developers had to be involved at all, even though all of their "no-code" tools introduced complexity that required developers to be involved as soon as you did anything more interesting than whatever was in the demo.
So yeah I made some money but I'd die before I was a salesforce dev again.
People in sales think it's pretty ok, and it's easy to find contractors who will set up or expand it for you. A couple of full-time devs could set up a CMS better tailored to your company with free components, but lots of places that use Salesforce don't have in-house developers.
Salesforce knows that its codebase is a hot plate of spaghetti, but it doesn't really matter because software developers "in general" aren't their target audience in any sense.
Yes, we did by 10,000 cores a year ago, no there is no capacity, yes we will run the ERP off an old win2k server, no we are not using 10k cores for seti@home.