Sunday ended up being less about code and more about figuring out how to talk about the code I've already written.
VirusMon
I've been kicking around the idea of doing a YouTube video on VirusMon, and today I actually sat down to work out what the angle should be. My first instinct was the obvious one — make it another "building a game with AI" tutorial. But the more I thought about it, the more that felt like exactly the wrong move. That space is already saturated with generic content, and frankly, it's not what makes this project interesting.
The real hook is that this is a solo dev shipping an actual multiplayer MMO for a hacker con. Not a toy, not a demo, not a 30-minute prototype. The AI-assisted part is incidental — the story is the constraint and the destination. That reframing changed everything about how I want to structure the script.
Rather than inventing a clean narrative arc after the fact, I went back through docs/progress/ — the screenshot diary I've been keeping — and the devlog.md. The idea is to build the script from what actually happened, beat by beat, instead of a tidy retrospective that glosses over the messy parts. There's a real pacing benefit to this too: the screenshots already give a natural visual progression, so the script can lean on them as B-roll instead of needing me to recreate anything.
Didn't finish the script today, but I've got the spine of it now: lead with the con deadline and the "one person, real MMO" framing, then let the devlog entries carry the middle. The AI tooling shows up where it actually mattered, not as the headline.
Good reminder that the story you'd want to tell isn't always the one that's actually true — and the true one is usually more interesting anyway.