The True Spirit of Vibecoding

The true spirit of vibecoding is about coding for outcomes, not technical perfection.

When I say outcomes, I mean the magic feeling of I want this to happen. I want my app that shows the weather in my city cross-referenced with outdoor gyms public and private, and I want each of those to have a weather score and a quality score. That's not a user story. That's not a specification. That's a want.

Embracing the vibe, the feel of it—I never said "I want a modal to sign up," or "I want this to be a button." Not at first.

When building vibecoder.date, I started with the landing page, and then the extension. I went by: what does a dating app need? Swiping. Messages. Profile. All of those things—not as a user story flow, but as abstract concepts that I then gave aesthetic to, and then UX.

Vibecoding is both additive and subtractive. You start by stacking blocks in the form of combining concepts: my app should do X and Y with Z. Then you take away what isn't working, add more.

It's a lot less formal than agile, waterfall, or any methodology. It's pure iteration. No theater, no ceremonies. That's why it's powerful—because ideas aren't versioned or snapshotted into work products (specifications, mockups, etc). Not by default. You can choose to use those as tools. I do sometimes.

If you take one thing from this: nothing is an obligation. That's why it works. That's why high agency and high discipline people excel without being made to give up the dream outcome.

Methodologies exist because scope was limited to the work of huge teams who didn't have the chance to vibe with each other. Even before AI, a small group of motivated individuals would tackle a problem naturally—with vibes, with connection.

With vibecoding there is another skill that you need, and that is vision beyond the constraints of tradition and corporate structure. You need to break free and embrace reaching for the impossible.

Impossible is something that must forever be what we seek to challenge. The four-minute mile was thought impossible. Space travel. Landing on the surface of Venus to take pictures. Impossible is never a no—it's a not yet.

I often joke about an AI model one-shotting AGI, but the truth is, we are closer than we have ever been to that. Even if the transformer architecture isn't the basis for it, the capabilities we have been able to extract from years, decades of research are now accelerating it even more.

The first time I had a phone that could play music well, I was happy. I could listen everywhere. When I bought my first Android phone, back in the Gingerbread days, the screen was 3.5 inches. The phone was tiny. I loaded it with music and movies and I'd carry a spare SD card with more. It felt like the future.

Nowadays even the most budget phones have massive screens and all the apps you could want. It's normal now. But back then, I could only imagine what would happen.

I think vibecoding has healed a part of me that I didn't know felt broken.

Knowing how tech works, being a technical person, really sets you up to think in constraints and viability. And if you work in the industry, you will develop a sense of it can't be done—not because it cannot, but because there is a barrier to it that you know the org or the team or even you cannot overcome.

That belief is almost always false. It's rooted in reasonable assumptions about viability.

It's very important then that we do not internalize this mindset outside of job roles. Vibecoding is permission to unlearn the limits you were trained to respect.

Dream big, speak out, and before assuming it's not viable or impossible, give it a try. You will be surprised at what you find.