On the skepticism of tech market hype
Every cycle promises that everything has changed. The fundamentals rarely do. A case for boring, durable engineering.
There's a rhythm to the industry now. A technology appears, gets a name, gets a conference, gets a wave of posts declaring that everything before it is obsolete. Money moves. Then the curve flattens, the survivors quietly ship useful things, and the cycle resets with a new name.
Hype optimizes for attention, not outcomes
The loudest claims travel furthest because they're built to. "This changes everything" outperforms "this is a modest improvement to a narrow problem" on every metric a feed cares about. None of that tells you whether the thing works, scales, or survives contact with a real deadline.
The fundamentals are stubborn
Latency still matters. Correctness still matters. The cost of a system is still mostly the cost of maintaining it, not writing it. No naming convention or funding round has repealed these. When a new tool genuinely helps, it helps along these old axes - it makes something faster, cheaper, or harder to get wrong.
Skepticism is not cynicism
Being skeptical doesn't mean refusing to adopt anything. It means asking for the boring evidence: what problem does this solve, what does it cost, what breaks, who's actually running it in production. The engineers who age well aren't the ones who chased every wave or dismissed all of them. They're the ones who asked good questions early and adopted late, on purpose.
Build things that would still be a good idea if nobody was watching. Most of the hype won't be. The work that survives usually wasn't trying to.