I was mentoring at a hackathon last year. A team of three was trying to build a real-time dashboard — auth, WebSockets, charts, the works. They had 72 hours and were already 20 hours in, stuck on wiring up a custom charting library they’d never used before.
I asked them: “What if you just used a pre-built chart component and spent those hours on the thing that makes your project yours?”
They looked at me like I’d suggested cheating.
That reaction is the LEGO problem. Somewhere along the way, we started believing that real engineering means building from scratch. That using someone else’s brick is a shortcut. That the value is in the novelty of the piece, not in what you assemble with it.
But LEGO sets don’t ship with raw plastic. They ship with bricks — standardized, tested, designed to fit together. The magic was never in manufacturing the brick. It was in the instruction book you threw away and the thing you built instead.
Web development works the same way. A
<Button />, a useEffect, a Tailwind utility class, an API route — none of these are impressive on their own. Everyone has the same bricks. The same React docs. The same npm registry. What differs is the architecture of the castle: which pieces you chose, which you left out, and how they fit together under constraints that are unique to your problem.Here’s the part most people miss: the hardest LEGO decision isn’t picking the right brick. It’s resisting the wrong one. The tempting abstraction. The “clever” custom hook that saves 4 lines today and costs 40 hours of debugging next quarter. The extra feature that makes the demo cooler but the codebase fragile. I’ve seen production systems buckle not because the pieces were bad, but because there were too many of them — each one reasonable in isolation, incoherent as a whole.
The best builders I’ve worked with share one trait: restraint. They reach for fewer bricks, not more. They pick the boring brick that fits over the shiny one that almost fits. They leave gaps in the design knowing that empty space is also a decision.
My take. I’ve been building things on the web for 7+ years now — side projects, hackathon MVPs, production systems at scale. The projects I’m proudest of aren’t the ones where I used the most impressive technology. They’re the ones where every piece earned its place. Where I could point to any brick and explain why it was there. That’s the LEGO mindset: not “how many pieces can I use?” but “what’s the fewest pieces that make this thing solid, clear, and easy to hand to someone else?” If you’ve ever loved LEGO, you already know how to build software. You’re just learning which bricks to leave in the box.