I Would Start Smaller

If I could go back to June 14 and sit next to myself before that first prompt, I would probably put a hand on my own shoulder and say, “Easy. One thing at a time.”

That first prompt was innocent enough. I asked the AI how we might build the greatest global soccer trivia game ever. Not “build it now.” Just give me a blueprint.

The answer was exciting. It sounded like Wordle meets Sporcle meets FIFA, which was exactly the kind of sentence that makes a person like me forget he has no coding background. I picked a “Stadium Night” look — navy and lime — and within about an hour I had a working trivia version on the screen.

That was the dangerous part.

Because once something works, even a little, my brain starts adding floors to the building before I’ve checked the foundation.

A few hours later I asked, “Should we build a fantasy league?” And that was the pivot. Suddenly this was not just a trivia game. It was a global soccer trivia and fantasy platform. Multi-league. Season-long. Head-to-head. Salary cap. Real match data through API-Football, which cost me $25 a month.

That was my first paid subscription for an app I had not launched.

I remember the feeling. Part excitement, part “am I really doing this?” It made the project feel more real, but it also made every decision after that heavier.

If I started over, I would not add the fantasy league that quickly.

I would start with trivia only. A smaller first version, but a better one.

Not a giant promise. Not “the greatest trivia game on the planet” on day one. Just a tight, fun soccer trivia game with enough depth to make someone want to come back tomorrow.

I’d spend my early energy on better questions. More of them. Different kinds. Real image rounds. Real audio rounds. Things that make the game feel alive instead of like 25 static questions sitting in a file somewhere.

That was one of the more sobering things the AI told me later. I kept asking if we had the best trivia game, the deepest knowledge, the strongest user experience. And the honest answer was basically: not yet.

It did not say that to be mean. It was just true.

We had the shape of something. We did not have the depth.

The fantasy side brought in a whole different set of grown-up problems. Accounts. Leagues. Match data. Payments. Legal questions. I even asked whether a paid-entry cash pool would count as gambling. The answer was complicated enough that I decided we would not touch the money directly and would deep-link to LeagueSafe instead.

That was probably the right call. But it was also a sign that I had wandered into a much bigger room than I meant to.

Even small bugs felt bigger because the app had become bigger. The first one that really got me was the fantasy account sign-up. I clicked “create account” and nothing happened. So I clicked again. Then I got red text telling me to wait 49 seconds.

It looked broken. It felt broken. I felt stupid.

Turns out the backend was working. An email confirmation had been sent. The screen just gave me no feedback, so I kept mashing the button until I hit the rate limit (basically, the app telling me to stop trying so fast).

The fix was simple: a loading spinner and a “check your email” message.

But that little moment taught me something big. When you add more systems, you add more places for a non-technical person like me to get lost.

I’m not saying I regret the fantasy league. Some of my favorite wins came from that side. Getting the head-to-head trivia duel to sync live between two players. Seeing a daily match render with a real-world fixture. Reworking the fantasy hero until it finally looked bold instead of sleepy.

And I still love that the AI could rename Top Corner to Hat Trick across the codebase in seconds. That felt like magic. UI, legal pages, app settings, translations — all of it. If I had tried that by hand, I’d still be finding old names next Christmas.

But if I were starting again, I would earn the word “greatest” slowly.

First, I’d build the trivia game I actually wanted to play every morning.

Then, once people were coming back for that, I’d add fantasy.

Small takeaway: excitement is useful, but it is not a roadmap.