The website I once felt embarrassed about is growing on its own.

Hey, I’m Jumani.

Before you visit the website, I want to tell you a little bit about me and how this whole thing started.

At the beginning of 2026, I was obsessed with one thing building something of my own.

The problem was, I couldn’t code. Not even a little. I didn’t know the C of coding.

At the same time, I was deeply into AI. I followed almost every AI update, new model launch, and major development. I knew what was happening across the AI space to the point where I used to joke that I was an AI buddy.

Somewhere along the way, I came across the term “vibe coding.”

I got curious.

I started researching what it actually meant, how it worked, what tools people were using, and whether someone with almost zero coding experience could actually build something with it.

I went deep into it.

Within about a week, I had learned enough about vibe coding that I felt ready to try building my own website.

But there was one problem.

There were so many vibe-coding platforms, and I had no idea which one I should use.

So I decided to try almost every one I could.

I started with Replit because I had already heard a lot about it. I tried it, but I wasn’t really happy with the results.

Then I tried Bolt.

Then Bubble, Emergent, v0, and several others.

For every platform, I tried to get as much as I could from the free plan because I simply couldn’t afford paid subscriptions.

But then I tried Lovable.

And for some reason, it clicked.

When I first used it, I was genuinely fascinated by what I could build with it. At the time, it was giving me a lot of things I wasn’t seeing together on other platforms — analytics, SEO, security, privacy, and many other things that were useful when building a real website.

And the biggest part for me?

I could do all of that on the free plan.

So I decided that Lovable was going to be part of my journey.

And that’s where the story of Halcyon begins.

Halcyon was a simple photo-editing website I built with vibe coding. It gives you 30+ presets that can quickly change the look of your photos without needing to understand complicated editing tools like Lightroom.

And as I mentioned earlier, I still hadn’t paid for any AI or development tool.

I didn’t have the money had no mentor,

So I built everything using free plans - prompting with ChatGPT’s free plan and building with Lovable’s free plan.

I was genuinely happy.
I had built my first website.

And the funny part?
When I started, I still barely knew how to code.

But building Halcyon pushed me to actually start learning.

I started learning coding stated from HTML,CSS while I was building the website yeah, I know weird way to start learning coding.

But it worked for me.

By August 2026, I had finished the first version of Halcyon and published it using Lovable’s free domain because I couldn’t afford a custom domain.

Then I started promoting it alone, still no mentor, no one else with me just me doing everything my own at the time, I had almost no followers on X still, I tried everything I knew.

I shared it in communities, Substack, on X, dev and anywhere else I thought real people might actually discover and use it.

At first, barely anyone came.
But slowly, people started visiting.
A few used it.

Then a few more.

Eventually, after a lot of effort, Halcyon crossed 100+ visitors from around the world.

And then came the part I wasn’t prepared for.

Some people started making fun of the fact that it was AI-built.People pointed out the Lovable branding.

Some of the feedback was useful, and some people were genuinely supportive, but the negative comments started getting to me.

Eventually, I stopped.

I went quiet.

For about a month and a half, I didn’t even look at the website.

I removed almost every trace of it from my profiles, stopped promoting it, stopped prompting, stopped marketing it and started posting about completely different things.

I basically abandoned it.

Then one day, I randomly decided to check the analytics.

And I was shocked.

The website was still getting visitors.

No marketing.
No promotion.
Nothing from me.

And somehow, it had crossed 200+ visitors from around the world.

People were still discovering it.
People were still using it.

And even more surprising, Google and ChatGPT were starting to recommend it.

I checked the analytics more closely.People were coming in every day.

I genuinely didn’t know what to think.Then it hit me.

I had spent all that time worrying about what people were saying about an AI-built website that I had forgotten to look at what the actual users were saying and while I was busy feeling embarrassed about it, the website was quietly doing something I didn’t expect it to do.

It was growing without me.

That changed how I looked at the whole thing So I started talking about Halcyon again not hiding the fact that it was AI-built not hiding the fact that I used Lovable.

Actually doing the opposite.
I started showing people the numbers and telling them:

This is an AI-built website.
I stopped caring about whether people thought that made it less legitimate. Because at the end of the day the website didn’t care who built it.

It just had to be useful enough for someone to visit, use and come back.

And that was probably the lesson I needed.

I thought I had failed because people laughed at how I built it.But while I was looking at the opinions, I completely missed the evidence.

The website was still alive.
People were still finding it.
And it was growing when I wasn’t even touching it.

So I came back to it.

I started working on Halcyon again, making improvements and bringing it back to where I wanted it to be.

I still have a lot to learn about coding, building products, marketing and everything that comes with creating something from scratch.

But now I look at Halcyon differently.It’s not the website I was embarrassed about anymore.

It’s the first thing I built when I didn’t even know how to code and maybe that’s the part I should have been proud of all along

Halcyon is here

Everything I did solo
Still learning Coding, building random things every day and figuring out what to build next.

still no mentor. still figuring it out. but i'm building again.