An engineer dad’s side project became a warm community journey

It started with finding toys for my daughters

When McDonald’s launched the Hello Kitty & Friends × Godzilla collection, I wanted to help my two daughters find the toys they loved. But each store had different toys, and useful information was scattered across Facebook groups, Instagram, Reddit, and friends’ messages. People kept asking which store had which number and when someone had last seen it.

On Friday night, I decided to use the skills I had learned as an engineer to make a simple, cleaner way to find and track reports. I worked through the night and launched the first version on Saturday afternoon: a map of more than 13,000 McDonald’s locations, nearby search, toy-number filtering, and a lightweight way to share what someone found.

From helping friends to a community data flywheel

The first version had a little over 200 reports from friends. I did things that do not scale: when I saw a parent asking for a toy number in a community group, I opened the map, took a screenshot, and shared it back. Slowly, reports spread from the Bay Area to Los Angeles and then to cities across the country.

People I had never met began sharing toys, sold-out stores, and local updates. Each report made the next collector’s decision a little easier. The site stopped being only a tool for my family and became a small act of mutual help.

A very real system-design lesson

The traffic grew faster than I expected. In less than two days, the free cloud-service quota was full. I was excited, but also very aware that this side project could not turn into an unexpected cloud bill.

The response became a practical system-design exercise: deciding which store data and historical reports belonged in long-lived caches, reducing unnecessary API and database reads, keeping map filtering responsive, and balancing freshness with cost. Concepts that usually live in architecture documents became decisions I had to make for real users every day.

The unexpected warmth of small support

On the third day, I added small support links for people who found the map useful. A few people sent $1 or $2 and left kind messages after finding a toy. The amount was never the point. It was the first time I had built, launched, and shared an independent product and received direct encouragement from people I did not know.

That recognition meant more than the money. It told me that the site was helping with a real, if small, problem.

What comes next

1. Keep improving search, map performance, and report clarity.
2. Make the architecture modular for the next toy collection.
3. Explore whether this community map model can help collectors in other countries.