I agree that is happening more often in the initial stages. And with a bit of experience/reputation you can command very favorable terms in the early stages of a start-up, at least in the present climate where seed type funding is relatively very easy to come by.

Or let's say you don't have much money and you want to avoid early VC due to loss of control and hefty dilution. In the US market, move to a college town (often quite safe), take a part-time $15 / hr retail job (CVS and Walgreens are both moving to that minimum in 2022; both are better than fast food and easier than Walmart type jobs), get a modest cost apartment ($650-$750), walk/bus/bike to work (no car expenses), keep your bills under $1k-$1.2k per month, build your start-up on Hetzner (new Virginia datacenter). Self-fund, build your thing, don't take early venture capital, maximize your leverage later on. Maybe occasionally do some side contracting work if necessary to fill up the personal treasury (10 hr per month or less).

The counter argument to that is primarily the speed angle, get bigger faster or get crushed by the VC funded crowd, and similar arguments. It likely varies by the case as to whether a given approach makes more sense or less (eg depending on what you're building, time to market may not be so much of the essence, or it may be critical).