Six years ago, I came up with an idea: a platform that could help people find plant-based products — not just food, but clothes, cosmetics, and more — all based on their location. A space where users could share discoveries, support local vegan-friendly businesses, and interact like a community: post, like, chat, and learn from one another.

I called it Vegan Check.

Without a tech co-founder or investment, I hired a development company. I did all the UX work myself — flows, wireframes, logos, research, testing. The app launched on the App Store and Google Play. Despite zero marketing, it organically passed 2,000 downloads. People were using it.

But I had to shut it down.

Every update—from bug fixes to button colours—came with a cost. Hosting, support, and app store fees made it financially unsustainable. I wasn't building a business but burning through savings, so I pulled the plug.

Enter AI. Enter Control.

Fast forward to today.

No funding. No agency. No code.

But Vegan Check is back.

I rebuilt the MVP using Lovable.dev, a tool that turns natural language prompts into real applications, and Supabase, which manages the backend, authentication, and database. Together, they became my new dev team — but this time, I wasn't just the designer. I was the builder.

With clear UX in mind, I prompted everything:

"Create a login page with Google and email."

"Allow users to post products with location, photo, and description."

"Make a public feed with like buttons and detail pages."

"Add profile settings and user dashboards."

And it worked.

Authentication, database integration, image upload, real-time updates — all life, all functional, and all built without writing a single line of code in the traditional sense.

But here's the truth:

It's Not Magic. It's Precision.

AI doesn't eliminate the need for product thinking.

It removes the barriers — not the responsibility.

You can't just type "make a vegan app" and expect brilliance. You need to understand the architecture, the flows, and the user edge cases. The AI won't give you what you need if you don't know what you're asking for.

I didn't code, but I knew how to think like a developer.

That's what made the difference.

Lessons Learned From This Second Build

• I now understand backend logic and database structure more than ever before — because I had to.
• AI made the process exponentially faster, but it still required dedicated days of work and problem-solving.
• I gained complete product control. No more dependency. No more invoices for colour changes.
• Building an MVP no longer requires €10k+ but involves clarity, design, and leadership.

What's the Future Now?

This raises a bigger question:

Will companies start hiring developers who don't code — but prompt?

Maybe. Maybe not.

But one thing is clear:

Prompt engineering, UX knowledge, and system thinking are merging into a new kind of builder that doesn't just type commands but understands context, user behaviour, data flows, and product value.

The real revolution isn't "no-code."

It's know-how and know-what.

And if you've got that?

You don't need to wait for permission or funding to build.

You need a prompt and a plan.

Want to check the Vegan Check?

Visit www.vegancheck.co

#AI #nocode #UXDesign #promptengineering #vegancheck #startups #indiehacker #supabase #lovabledev #productdesign #MVP #buildinpublic