Help! My Teenage Son Hijacked My Newsletter...

Hi, and welcome to The Atomic Builder!

This is where product managers, founders, and non-tech creators learn to master AI-powered software building…

BUT HOLD UP…THIS WEEK…

…my son has hijacked the newsletter. I’ve got my feet up, and he’s here to tell you about his recent adventures in building with AI.

Take it away, Ihsan!

Introducing Ihsan

Hey, I’m Ihsan, and I’m a 15-year-old computer science student who’s all about tennis and gaming. My dad’s always going on about AI and building stuff without coding, so instead of rolling my eyes one more time, I thought I’d see what all the fuss was about.

I decided to make a few mini-games using Bolt. My first attempt? A Whack-a-Mole game and a homepage to start a mini-game collection. It was fun, but way harder than I expected! Here’s how it went down.

Click here to play the Atomic Arcade.

Why mini-games?

I’ve always loved gaming, but let’s be real - making a full video game sounds impossible when you’re just starting out. Mini-games felt more manageable. They’re quick to play and (usually) don’t need crazy designs to be fun.

I started with Whack-a-Mole because I thought, “How hard can it be?”

What did I actually make…

Game 1: Whack-a-mole

Building Whack-a-Mole was my first real dive into making a game, and it seemed simple enough at the start. I mean, it’s just moles popping up, right?

WRONG!

Once I got into it, I realised there’s so much going on behind the scenes. I had to figure out how to make the moles appear randomly, keep track of scores, and even add a timer to make it more challenging. The best part? Seeing it work. The hardest part? Getting the timing just right so it felt fun but not impossible.

Wow. This was the first thing I ever made!!

Game 2: Maze Runner

Maze Runner was a bigger challenge. The idea was to guide a character through a maze without hitting the walls. Getting the character to move smoothly with the arrow keys was tricky. I also wanted to make the mazes harder over time and even tried adding moving obstacles like Pac-Man ghosts.

But let’s just say… ghosts weren’t on my side, and the glitches won this round.

Game 3: Tic-Tac-Toe

Tic-Tac-Toe was probably the easiest game to create!

I still ran into problems though. I wanted to make it fun for two players while also adding an option to play against a computer. This meant learning how to code a simple AI to make moves for the computer.

Getting it to make its move quickly but not too quickly was one of the most important parts for me! AI helped me brainstorm how to structure the game logic, especially for checking who won after each move.

The games you’ll never play

Not everything went smoothly. Here are the two games I couldn’t quite finish.

Platform Jumper

This one was tough. I couldn’t get the character to stop falling through the platforms, no matter how many times I tried to fix it. Gravity was definitely not on my side!

Space Invaders

This game got further along, but it wasn’t quite there. The biggest issue? Only one wave of ships would appear. I wanted to add scoring, tougher waves, and different types of ships, but I couldn’t get it to work right. Calling it annoying would be an understatement.

After two frustrating attempts, I decided to switch gears and work on Tic-Tac-Toe instead, which AI seemed to be better at.

Final Thoughts

This whole experiment was a lot of fun, and it helped me learn about the work that goes into even the simplest games. I’m glad I’m studying computer science - it’ll definitely help me understand this stuff better.

This Week’s Build Beats 🎵

Each issue, we pair the newsletter with a track to keep you inspired while you build.

This week, because AI building is all about levelling up.

🎧 “Eye of the Tiger” – Survivor

Grab the playlist on Spotify - I add to it each week!