↓ Skip to main content
The Drew Archive One house, many rooms
The Terminal · · 4 min watch

My Old Laptop Borrowed My New Mac’s AI to Make This Game

My 2019 MacBook Pro used OpenCode with a local Qwen model running on my M2 Max over a private Tailscale connection to build Neon Asteroids.

Author Drew
4 min watch
Watch on PeerTube Watch on YouTube

My 2019 MacBook Pro is a useful Omarchy workstation, but it cannot run a large AI model on its own. Instead, OpenCode sends inference requests over my private Tailscale network to qwen3.8:27b-mlx, running through Ollama on an M2 Max MacBook Pro with 96 GB of unified memory.

This short demonstration shows that two-machine workflow in practice and the result: a browser game called Neon Asteroids. The game is fun, but the more interesting part is that the coding agent ran on one computer while the model powering it remained on hardware I own somewhere else on my network.

Read the full technical story for the architecture, setup, development process, and the bugs I encountered along the way.

Play Neon Asteroids or read why Tailscale has become foundational infrastructure for the systems I run.

Where to go next

Keep moving through the archive

This piece works best as part of a larger path. Pick the next doorway that matches how you want to keep reading.

Stay in this realm The Terminal Browse more writing from this section if you want to keep exploring the same part of the archive. Read the adjacent piece What Happened to Roosh V? The story of Roosh V's transformation from a notorious internet personality into an Orthodox Christian who renounced his former …

Keep exploring

Explore beyond this room

These are the closest nearby pieces in the archive, with extra weight given to the same realm and trail when possible.

The Terminal Video

AI Help with an External Mic in Omarchy

A quick external-microphone and screen-recording test that shows how Omarchy's built-in AI can help configure and operate a Linux computer more quickly.

Comments

Join the discussion. Be thoughtful, stay on topic, and treat other readers as neighbors. Email sign-in and comment data are described on the privacy page.