The Vibe Coder's Manifesto

I'm not a developer.

I say this not as a disclaimer but as a statement of fact. I never learned to code properly. I can read JavaScript well enough to debug something obvious. I can write SQL queries that mostly work. But I am not, by any traditional definition, an engineer.

And yet I've built nine products.

The term people use now is "vibe coder." I think it was coined semi-pejoratively—developers making fun of people who use AI to write code they don't fully understand. And fair enough. There's something genuinely dangerous about deploying code you can't debug.

But there's also something powerful happening. For the first time, the bottleneck for building things isn't technical skill. It's taste, persistence, and problem clarity.

Here's what I've learned in a year of vibe coding:

AI doesn't replace understanding. You still need to know what you're trying to build. You still need to break problems down into components. You still need to debug when things go wrong. AI is leverage, not magic.

Start with the problem, not the solution. The clearer you are about what you're trying to solve, the better AI tools work. Vague prompts produce vague results. "Build me an app" produces garbage. "I need a way to show my grandmother her schedule on her TV without her needing to log in" produces something useful.

Ship ugly. The first version will be embarrassing. The second version will be less embarrassing. The tenth version will be something you're actually proud of. But you never get to version ten if you're waiting for version one to be perfect.

You can't automate taste. AI can generate options. It can write code. It can create designs. But it can't tell you which option is right for your specific users with your specific constraints. That's still on you.

I don't know what the future of "vibe coding" is. Maybe the tools get so good that the distinction disappears. Maybe we're in a weird transitional moment that won't make sense in ten years.

But right now, if you have a problem you understand deeply and the patience to iterate, you can build things. Without a computer science degree. Without venture funding. Without a team.

That feels like a genuinely new moment.