With AI, The Real Edge Is Knowing What to Build

17 Jun 2026 #coding

“But… how did you know what to write?”

That’s what Greg Baugues’ daughter asked him after her first encounter with vibe-coding. She was curious if there was a time when we had to type symbols, one by one.

It isn’t about what symbols to type

Coding used to be about the craft: Mastering symbols and perfecting the lines of code. If you knew what to type and could make it work, you could find your way as a coder.

But AI has changed our job description by making code cheap. Anyone is just a few sentences away from working code.

Knowing what symbols to type isn’t the edge anymore.

It’s about what to build

As I write this, I’m working on a legacy migration for a hardware shop.

Its main page is an entangled mess of over 40K lines of Visual Basic code. Migrating it to anything newer is time-consuming.

To ease the pain of migrating it, management suggested starting some refactoring work. The migration has dragged on for years, still far from completion. The refactoring work won’t show any visible progress on a late project.

Prompt an LLM to migrate it to Blazor isn’t hard. But its output will only create more problems, without the mental models of the codebase. The challenge is deciding the right approach: refactoring, migration, or rewrite.

The real edge is: knowing coding is solving problems and choosing what problems to solve.

When coding is cheap, your edge comes from outside your IDE. To help you grow beyond syntax, check Street-Smart Coding Manifesto.