The First Wall

I think the first wall hit me harder because the first few minutes had been so exciting.

On June 2, I gave the AI one enormous prompt. Looking back, it was almost comically ambitious. I basically asked for an entire philanthropic ecommerce and community website for Beads of Hope T1D in one breath: shop, custom products, community, checkout, the whole thing.

And then, about 20 minutes later, there it was.

A real-looking storefront.

Not a sketch. Not a vague idea. An actual shop page with categories and products and a checkout that seemed to be wired up. The AI had created database tables, Google login, Stripe checkout, and more things I did not fully understand but could tell were “serious website things.”

I remember feeling thrilled and completely disoriented at the same time. Like I had opened a door I did not know existed, and then immediately realized I was standing in a room full of controls I could not read.

Then the first wall arrived that same evening.

None of my uploaded product images were showing on the published site.

Not some of them. None.

On top of that, the shop page had duplicate category buttons. And every time I clicked the empty box where a product image was supposed to be, the screen jumped back to the top of the “all” category.

It felt like the site was mocking me a little.

I told the AI, “images are still not visible and each time I click on the box the image should be in, the screen immediately reverts back to the top.”

That sentence pretty much captures my mood: confused, annoyed, and trying very hard to sound useful.

This went on for hours. The AI would try something. I would publish or check again. Still broken. Another attempt. Still broken. The images were supposed to be the whole point. Beads of Hope is about handmade bracelets, necklaces, badge reels, and pens. If people cannot see the products, there is no shop. There is just a very polished empty shelf.

That was the moment I felt the gap between “the AI can build something fast” and “I understand what is happening.”

Because I didn’t.

I didn’t know if the problem was with the images, the product listings, the published site, the database, or something else entirely. I didn’t even know what questions to ask at first. I only knew what I could see: blank product boxes, duplicate buttons, and a page that kept snapping back to the top like a rubber band.

A few days later, I hit another version of the same feeling when the customize page threw a TypeScript error. TypeScript is a coding language that tries to catch mistakes before the site runs. That sounds helpful in theory. In practice, the error said it could not find something called @/assets/blue-ribbon.png.asset.json.

I stared at that like it was written on the side of a spaceship.

I knew what a blue ribbon was. I knew what a PNG image was. I had no idea what an asset.json file was, and honestly, I still barely do.

The funny thing is, some of the things that stressed me out most were tiny design details. Moving a blue ribbon down a quarter inch. Making the “T1D Hope” text stand out better with white shading. Getting a hover zoom on badge reel photos, then deciding 3X was too much, then 2X, then only wanting it on badge reels.

Tiny things. But they did not feel tiny.

They felt like the difference between “this is a real brand” and “this is someone’s weekend craft project that got loose on the internet.” Maybe that is too harsh, but that is where my head was.

The first wall was not one clean problem. It was the feeling of being out of my depth while also caring too much to walk away.

I was proud. I was frustrated. I was embarrassed that I didn’t understand the error messages. I was also stubborn, because this site was not just an experiment to me. It was for Type 1 Diabetes awareness. It was for my son.

So I kept going, one broken image and one confusing message at a time.

Small takeaway: fast progress is exciting, but the first broken thing is where the real learning starts.