On a per employee basis, the overall likelihood adjusted expected value and liquidity of most startup shares generally underperforms the market, almost to zero if biased towards liquid earnings like it should be. If you end up the former employee of one of those startups (which is pretty common), you end up in a place where even if the company is booming 5, 10 years later, no liquidity event has happened nor is in sight for employees. If you ask if you could give someone $9-19M liquid evenly split over the next two decades or several tranches of $1-100M potentially forever illiquid stock and $2M liquid, which one do you think they would pick and should pick? Even if the max payout is lower, the mean payout is orders of magnitudes higher and creates access to bootstrapping. I'm not going to say you made the wrong one for obvious reasons, but do you think it's smart to attach your lifetime net worth to such illiquid assets?
Garry, you entitled your blog post "Working for Microsoft cost me $200 million" but you neglected to mention that you had debt you were paying off. For many other people, that blog post could easily have been called "Why working at startups instead of Microsoft cost me $20M and left me in the debt I started in". That debt neatly encapsulates one of the societal problems with student loan debt. It creates a caste system. The less generational wealth you are born into, the more your stepwise freedom in any direction is piece by piece restricted. You can't make choices that people born just a bit wealthier than you could make because they're cost step functions.
It would indeed have been a poor decision of you to take that Palantir job out of college with your financial situation. That smart decision would have been to keep pursuing the traditional big company engineering promotional path. Any action you take that deviates from that path towards a sustained lower liquid compensation on a year over year basis is one where you're taking a hit to liquidity by paying opportunity cost to make a default-illiquid investment (or multiple of them over time). If you combine all the post-tax income you've lost with the compounding interest it could have made, it's substantial on a career-long basis.
I would advise folks to work at startups for the same reason I'd advise them to work at a fashion magazine or work at a record label intern. It still has some cachet as culturally creative labor, and you can certainly make a living in the industry, but it's not a great place to stay long-term unless you're independently wealthy, are running the show, or you found a firm cleanly in hypergrowth and manage to grow with it. The antics and financial shenanigans take a toll on you. You can't rely on it as a sane vehicle to consistently build wealth, because it's structured like a swap heavily in favor of the company's investors, and you're footing the bill with your labor in exchange for an IOU that the company will IPO.
Tens or hundreds of millions of extra income and RSUs from their big company paycheck could be going straight into their nest egg and compounding over time, making it viable for them to make a major life decision without hardship. It's reckless and irresponsible to recommend others to engage in without a significant buffer of pre-existing wealth, and with the expectation that (like many other kinds of investing) all of participation is at risk of full loss.