That's a lot of equity if the startup treats that lost income as if the engineer had invested that amount in the last fundraising round. Options instead of stock, common instead of preferred shares, and risk adjustment should all probably increase the equity amount. I'm not sure that most startups would give mid-level engineers millions in equity.
If that engineer instead works for 100k and takes 100k in illiquid equity which may or may not be worth something in 10 years, they are losing well over a million dollars in opportunity. Given that 5 more years at a top-tier firm would double their compensation they could easily be giving up multiple millions. To be employee #5-50 this individual should be getting an equity option that would reasonably be worth millions in an exit scenario for it to make sense.
Most startups simply aren't competitive for this mid-level engineer given current compensation practices. Which leads to a sub-selection of talent for those who for various reasons aren't willing or able to work at a firm willing to pay that level of compensation. Bear in mind, if the engineer had liquid equity from a public corporation they would be able to leave equity in hand after 4 years.
Whether this is a problem or not is an interesting question for startup CEOs. Generally I've found success in software is more closely correlated to the quality of employees than the quantity, and historically companies with generous compensation practices seem to be the successful ones.
This is the tricky part, especially if you are very early.
Really what you want is the (statistical sense) expected value of the equity be something like the opportunity cost, for it to make sense for you. But until the company has been around for a while and through a few rounds, valuation is mostly a fantasy...