Bringing AlphaGo Zero to a Simple, Addictive Game

Hey everyone!

Today, I'm excited to share something truly special — a small game that started as an experiment and turned into something much more impressive than I expected.

At first glance, it looks like a simple twist on Tic Tac Toe.
But behind the scenes?
It’s powered by the same deep reinforcement learning algorithm used in AlphaGo Zero — the AI that changed the world of artificial intelligence forever.

How it all began:

I wondered:

Could we take one of the most powerful AI learning techniques ever created... and apply it to something incredibly simple — yet smart and fun?

I didn’t use a massive server farm or endless computing power.
I trained a lightweight version of the AlphaGo Zero algorithm on just a Mac mini.
Through self-play, it learned from scratch — just like AlphaGo Zero did — no human strategies, no shortcuts. Pure learning.

The Game Twist:

It’s not just standard Tic Tac Toe.
In this version, you stack stronger units on top of weaker ones, opening up a whole new layer of strategy.
Every move feels fresh — do you block, conquer, or set a trap?

• Stronger units conquer weaker ones.

• Run out of a unit type? You’ll need to adapt.

• Win the classic way — three in a row — but with much deeper thinking.

It’s simple enough for anyone to pick up, yet deep enough to surprise you every time you play.

Why I'm Proud of It:

• AI Intelligence — built on real AlphaGo Zero techniques.

• Training on a personal Mac mini — proving that innovation doesn’t need huge resources.

• A simple, smart, addictive twist — making every match exciting and new.

Ready to challenge yourself?

If you love games that are easy to learn but hard to master — and if you’re curious about how powerful AI can transform even the simplest ideas — this is for you.

🚀 Play now and see if you can outsmart an AI born from AlphaGo Zero!