Product Vibe Designing
When "Vibe Coding" Stopped Being a Joke
The first time I heard the term vibe coding, I rolled my eyes. Hard. It sounded like one of those Silicon Valley buzzwords that belong on a hoodie, not in a serious workflow. But then people started using it in earnest: large language models (LLMs) that take your intent, your "vibe," and turn it into working code.
At first my reaction was pure cynicism:
Sure, let’s see what amazing products you can ship with that. Maybe add "vibes" as a skill on LinkedIn— endorsed by your dog.
Then cynicism softened into skepticism:
Okay, code is text, so it makes sense. But there’s no way you can build a real, usable app this way.
Then came curiosity. And then, inevitability.
The Acceleration Effect
Once I gave LLMs a real place in my workflow, the change was dizzying. I went from idea to working MVP (minimum viable product) in days, sometimes hours. A few iterations on a database design and schema that used to take a week? Now done before lunch. A working full-stack app? Spin it up in a weekend.
The dopamine hits were real. Shipping has always been my fuel, but now the frequency dial is turned to 11.
I also realized: this works best for people with a certain skill mix. I have just enough coding background, product leadership, and architecture experience to see both the limitations and the magic. The tech doesn’t replace expertise; it amplifies it.
From Shipping Code to Shaping Design
The surprise upside of this acceleration: time. With LLMs covering ground I used to sweat over, I can invest in skills I always wanted but never had bandwidth for.
Design has always been that itch. I know good from bad design, but I’ve never been great at it. In the past, it felt smarter to double down where I already had strengths—coding, architecture, product strategy. Design was always the "maybe someday" skill.
Now I’m experimenting. I feed wireframes, PRDs (product requirement documents), and text prompts into AI design tools. The results? Mostly disappointing. Lovable and similar platforms can spin up a decent landing page, but they all feel cookie-cutter. Different wrappers, same vibe.
Maybe that’s because vibe designing isn’t really here yet. If vibe coding was v1.0, maybe vibe designing is still just a next step on our collective roadmap.
Why I’m Still Learning Design the Old-Fashioned Way
Here’s the kicker: I don’t think skipping design fundamentals is smart, even if AI tools eventually nail it. When vibe designing arrives, it’ll be like vibe coding: a massive amplifier. If I know the principles of proportions, color theory, and composition, then when the tools get good, I’ll be in the same sweet spot I am with coding now.
So I’m picking up design the slow way—and I’m enjoying it. Sketching, studying, comparing. Not because AI won’t get better, because, oh yeah, it will. But because I want to be ready when it does.
Lessons Learned
- Buzzwords are annoying—until they’re not.
- AI tools don’t replace expertise, they multiply it.
- Speed creates space. What you do with that space matters.
- Cookie-cutter designs highlight the gap between automation and originality.
- Investing in fundamentals now means bigger leverage later.
Next Steps
I’ll keep experimenting with AI design tools, but I’ll also keep learning the craft itself. If vibe designing takes off, I want to be someone who doesn’t just push the button—I want to shape what comes out of it.
And honestly? It feels like an amazing time to be alive
— Builders and dreamers, the future is very much hiring.