The real trick is to invest in something you do have access to insider information about and that you do do full time: being yourself. Invest in human capital, aside from a financial baseline to diversify for when your human capital starts to rapidly depreciate.
In some ways this is cheap. Reject certifications besides the most basic (a degree). Focus on getting an education instead. Buying a top of the line computer, high speed internet access, and books to learn from, for developers, is an almost negligible expense that will go a long way. And the most valuable way to learn--creating useful projects--can more than pay for themselves.
In other ways this is expensive, though. Choosing to invest in a 401k* takes an hour or two of your life per year, while investing in yourself is at least one or two hours a day. This is very, very expensive if you're already spending 50 hours a week sitting at a computer pounding away at brain-deadening code. (On the other hand, if you can get paid while also increasing your human capital... you've hit gold. Stay there until you've stopped rapidly learning, and then jump to the next big thing.)
Key point, though: there are multiple areas you have to invest in. Yes, a professional skill like coding is useful. But who knows what it'll be like in 5 or 10 years? Make sure to put time into your relationships, your physical health, your non-coding hobbies (drawing, banjo, typography, whatever). This diversification exposes you to more long-term investments: they help you maximize your luck surface area and will come in handy surprisingly frequently down the line.
How do all these retire-at-30 articles fit in? They aim to maximize financial investments early on by under-investing in many categories of self-capital, hopefully catching a good bull market, and rapidly switching to building self capital at 30. The obvious flaw is the assumption of outsized returns--they're not going to be as big as hoped--but that can be dealt with by tinkering with the numbers a bit.
The more fundamental issue is that it's not diversifying. It's risky. If you've saved up 500k by 30 by working long hours and frugally cutting coupons on your time off, that's nice, and if things work out right you might be fine. But suppose the defaulting of some government thousands of miles away sets off a chain reaction of bank failures that ends up massively contracting the economy you live in. There goes most of your savings. Yes, you might have invested in bonds, but you wouldn't have been pulling in those massive returns you were banking on to retire so early. And your job, having been funded by massive amounts of loose capital, suddenly disappears. Oh, you're farked, and you'll have to start nearly from scratch again after the economy recovers. Back to giving up your weekends to the whims of an MBA. (If anyone's willing to hire an expensive 35-year-old developer when there are all these recent college grads willing to work like dogs so they can retire at 30.)
Or even simpler: you hand in your resignation on your 30th birthday, walk out the door, and are hit by a semi driven by some overworked and drugged up trucker. Wow, that sucks. At least all those hard-earned dollars will go to some charity or another.
The ideal, I think, is to semi-retire as soon as you can, and work 15 to 20 hours a week at jobs you find interesting or fun. You get the best of both worlds and have diluted the amount of risk you face at any one time.
*Controversial statement here: 401k's are the biggest scam alive today, you're not only freezing your capital but also betting on taxes being lower in the future than they are today.
