I do hear sometimes from programmers who are kind of sad that they don't have the opportunity to write game engines from scratch like I did and have it matter or make an impact...
I was definitely a programmer who felt like I'd missed a golden era of opportunity. I didn't really start programming until my 30s. Just like many devs hoping to have an impact on the industry, it's sometimes felt like the golden era has already passed.
Surveying the market from a more entreprenurial mindset and looking for a potential niche to fill with a SaaS app (or worse, a social network), the situation appears even more dire! News aggregators have been done and Reddit is thoroughly dominant. The professional social network is done and LinkedIn owns it. Market after market is done. Everything in tech is done.
# Technology doesn't stay technology
Part of the problem is the word "tech". There are two different—but overlapping—ideas that are both commonly referred to as "tech".
By its literal definition, techonology is anything that involves applying scientific knowledge for practical purposes. In a literal sense, we're surrounded by tech. Some, such as the spoon are ancient and others, like the Kinesis keyboard I'm typing on are much newer.
But people don't talk about ancient inventions like the fire bow, or Clovis points as "technology", outside of a historical context. Most people don't even call somewhat modern inventions like light bulbs, refrigerators or microwaves as technology either. When we call something technology, we generally mean something invented recently... or we mean something related to computers.
Douglas Adams summed it up into three rules:
- Anything that is in the world when you’re born is normal and ordinary and is just a natural part of the way the world works.
- Anything that's invented between when you’re fifteen and thirty-five is new and exciting and revolutionary and you can probably get a career in it.
- Anything invented after you're thirty-five is against the natural order of things.
Having a musician turned firmware hacker and then programmer for an uncle, I'm a bit more optimistic about aging and see the boundary as more about whether a given technology has been widespread for a decade or not.
Cassette tapes and microwave ovens used to be fantastic new technologies in the late 60s and early 70s. Now they aren't. DVDs, the web and 3D video games used to be amazing technologies in the 90s. Now they're taken for granted. Smart phones were incredible in the late aughts. Now they're just phones.
# When pundits say "tech" they generally mean computing
The rapid advances in personal computing during the 80s, 90s and early 2000s was nothing short of breathtaking. Computers were gaining new powers every year or two and PCs were deprecating nearly 50% per year! The web took the boom to an exponentially growing number of users. 
It's entirely natural that computer software dominated the discussion of technology for a generation. First journalists, and now many of us are referring to everything from web SaaS apps to mobile productivity apps as "tech". Since the wider meaning of tech hasn't been abandoned, this makes it very easy to conflate the two.
But not all technology is computing and not all computing is technology anymore.
# Where is the edge?
To an ambitious engineer looking for a market opportunity in "tech", things look crowded. Advances in programming languages, frameworks, open source collaboration, dev ops, visual site-building tools and more have made it dramatically easier to create both web apps and mobile apps than it was a decade ago. This has lead to a flood of competition and an increased importance of marketing and other skills an engineer might not have.
But web apps aren't (necessarily) tech. They're "tech".
But not all is lost for the aspiring engineer.
There are newer technologies, closer to the frontier that aren't common in in 2020 but will be in 2030. Some of them are still wide open to people with more learning power than earning power. Bio-tech, blockchain, and VR are tech and the technical challenges to be solved are real. It's a lot harder to ship a product in those areas than to build yet another SaaS targeting small to medium sized businesses. But that's a good thing if technical skills are your strength.
For those whose strengths lean more to the marketing, sales or design sides, the B2B SaaS may represent a better goal. This is good for everyone. There's vast landscape for different people with different comparative advantages to find their niches.
# You are not too late
...here's where some perspective really helps - I can remember when I was a teenager, I thought I had missed the Golden Age of 8-bit Apple 2 gaming, that I was never going to be Richard Garriott...time went by, and I got to make my own marks in things after that. And, in that time, I also see so many opportunities that have come by.
The 90s PC wave was great - I was happy to be there, and I'm glad I took a swing and knocked one out of the park with that. But since then, we've seen mobile games, and web games, and free-to-play games, the Steam revolution...and now virtual reality. And all of these are amazing!
So, yeah, the opportunities that I had aren't there for people today - but there are new and better ones. And personally, I'm more excited about these than anything that's come before. So, thank you very much for this honor, but I'm just getting started.
-John Carmack (BAFTA acceptance speech)
The speech John Carmack gave when accepting his BAFTA fellowship filled me more professional optimism than I've felt in quite a while. Even he felt like it was "too late" because of missing all the opportunities of the 80s.
In retrospect, that seems crazy. The web was huge and just around the corner.
 Well more of an S-curve, but it looked exponential then.
 As one commenter on HN pointed out, it's a spectrum. In Uber's early days, its technology was crucial to its business. Over time, they'll gradually become less of a tech company if they coast and stop innovating.