This economic situation is really a kind of hybrid of full-salary and apprenticeship, where you're partially compensated in startup experience. I'm doing it now so I can learn the ropes of an early-stage startup while paying off my student loans. But long term I suspect your EV is better as a founder or cash-compensated so long as you properly invest your cash compensation.
Even better is to be one of those VP's who come in during Series C at an already-successful company and somehow make a market-salary and get 10%+ in equity. You know, the fuckers in suits who come in and do nothing.
The startups presenting at Workatastartup range from established companies 5 years old to startups from the most recent YC cycle that are currently run by single founders and are looking for people to be de facto cofounders, in every sense including equity.
As to your point about joining a post-YC company, employees #1-#5 are generally defacto cofounders and are often given similar equity. I'm not disputing that. I'm talking more about (roughly) employees #6 -> #30. At that point you have a product vision, but no concrete product. Since the major work of a startup is discovering the details of what to build, a majority of the work isn't done yet, and the real risk hasn't been mediated. However, a majority of the equity has already been passed around.
The risk and reward are certainly correlated. To say for certain you'd need a rather complex economic analysis that factors in opportunity costs to really say what the monetary sweet-spot is on the founder<->enterprise-employee spectrum. My suspicion is that the startup talent market isn't very efficient; As an industry we haven't really found a good way to judge technical skill other than working alongside someone for months. This hurdle means all the good jobs come via social connections. All the good positions are taken by the founder's network, and all the great devs already have a good gig.
I think this is true of the hiring market in general. "You made $80k at your last job as a senior developer? Super. Alright, we have to ask: FizzBuzz. Can you do it?"
What's the harm in asking if it's a quick and simple discriminator?
I don't think this is true, at least not for consumer internet startups. If you can't build a concrete product in this space with 5 people, you probably can't build one at all.