I Became a Better Cook Through Vibe Coding

I’m not a software developer.

I’m technical, yes, but I’ve never written production-level code as part of my work. Still, curiosity got the better of me after hearing a podcast episode about vibe coding. I decided to try it for myself.

What followed surprised me.

Within a few hours, I had a working script that scans and optimizes my personal recipe library of over 100 dishes. From there, it evolved quickly. I added a cuisine progress tracker, suggestions for missing “essential” recipes per cuisine, and an automated analysis of my cooking skills and knowledge gaps.

The end result: roughly 5,000 lines of working code. Written entirely by AI. I didn’t write a single line myself.

Is the code elegant or perfectly optimized? I honestly don’t know. And that’s not the point.

The point is that I’m not a programmer, yet I was able to build a functional application that solved a real problem for me. This wasn’t a demo or a toy example. It changed how I cook. It pushed me to practice more deliberately. It even shaped what I’ll focus on next. This quarter, I’m diving deep into Indian cuisine.

This is still “just” a hobby project. But the implication is bigger.

If you are a software developer, your role is already shifting, or will soon. The work moves away from writing code line by line and toward steering, reviewing, and validating what AI produces. The same shift is happening far beyond software development.

AI doesn’t just help us optimize existing processes. It expands who gets to build.

We’re only at the beginning of this LLM era. If this is what’s possible today, it’s worth paying attention to how quickly agency, creativity, and execution are converging.

Sometimes, better tools don’t just change how we work. They change what we choose to work on.