The Rise of Small Software Companies

A few years ago, building software was expensive, slow, and bureaucratic. Not anymore. Today, a capable team can build in a week what used to take months, endless meetings, and massive org charts. LLM-based tools—Claude, Cursor, and others—didn’t just speed up development: they broke the rules of the game.

For decades, large software companies won through scale. More people, more budget, more processes. But that logic has become obsolete. The same structure that once provided an advantage is now an anchor. Slow decisions, bloated teams, layers of approval that can’t survive a market that changes week by week.

Meanwhile, the cost of producing software has collapsed. An MVP that used to take three or four months can now be created in days. And no, we’re not talking about a simple prototype: often it’s better, cleaner, and more usable. Not because people are smarter, but because the tools brutally amplify those who know what they’re doing.

Here’s the key point: the bottleneck is no longer writing code. That problem is largely solved. The new bottleneck is something else: thinking well. Defining the right problem. Choosing what to build and what not to. Having judgment. LLMs don’t fix bad ideas, but they turn good ideas into real products at enormous speed.

And this favors, almost by design, small teams. Less internal politics, less noise, fewer excuses. An idea is discussed, implemented, launched. You see what happens. You correct course. You repeat. Short iterations, fast learning, no friction.

We work this way: a small team—two developers and a UX/UI designer—obsessed with speed and precision. We don’t compete over who writes more code (that’s no longer the problem!) but over who makes better decisions per week. That’s what works today.

Small software companies won’t win through brute force. They’ll win because the game has changed. Because in a world where code is no longer the problem, mental clarity and fast execution are worth more than any giant structure.