Vision

I'm building a game studio that fits inside one app. One human director (you), a crew of AI agents that do the actual building, and a platform that takes you from "I have an idea for a game" all the way to a real, finished game for sale on Steam. In any genre. Without writing a line of code, drawing a single sprite, or recording a note of music.

I'm building it because I want it to exist and I can't believe it doesn't yet.

Here's the thing I keep coming back to: making a real game, alone, is brutally hard. Not because any single piece is impossible, but because a game is a thousand decisions that all have to agree with each other; the art has to match the music has to match the mechanics has to match the story. In a real studio, a bunch of specialists hold all of that together in their heads, and every one of them delivers skilled outputs in their domain. But a solo dev has to be every one of those specialists at once, and that's a monumnetal load - one that most of us just can't hold alone. So MOST amazing games get imagined and never get made. *The mission is to release the potential.*

The robots are genuinely good now at every individual piece (the code, the art, the music, the writing). What nobody has really cracked is the part that actually matters: getting all of those pieces to work together, coherently, around one person's vision, at the scale of a whole complex game. That is the hard problem, and that is the thing I'm building.

The way it works is that your design lives in one place that everything else answers to. Your pillars, your mechanics, your art direction, your characters, your world, all of it, captured as you direct and shaped however you want, from big sweeping calls ("make the whole thing more tense and lonely") down to tiny specific ones ("the rate of hunger decay is a bit too slow" or "the range of the polearm is too much"). The art agent reads it. The music agent reads it. The engineer agent reads it. Nobody is handing off across a gap, because there is no gap. You're the director, they're the crew, and your design is the thing that keeps everyone in sync.

And to be clear, this is NOT "type a sentence and get a game." That's the vibe-coding thing, and it spits out generic slop that isn't really yours. You do the designing. The robots do the building. You stay the author the whole way through.

What I'm aiming at first

The first version has a simple, ruthless bar: take an idea and turn it into a real, functioning, genuinely complex 2D game you'd be proud of, end to end, through the app alone. I'm keeping everything else skinny enough to actually hit that, then making it deeper once it's real. A lot of the fancier stuff (sound effect generation, deep animation pipelines, AI running live inside the shipped game, all the launch and live-ops machinery) comes later, layered on top of the same core.

I'm building it in public

Out loud, with all the robots, including the parts where I'm wrong and have to back up. The progress matrix on the home page tracks every piece from research to 1.0 and updates the day something lands. Posts go out by RSS and email the same day. No funnel, no paywall, no "drop a follow if you agree." If you want to watch me either pull this off or fail trying, this is the place.

Where this goes

If I can make it work, I want it to be where a serious indie builds the game they actually have in their head, with leverage they could never get otherwise, without giving up authorship or taste or the right to make the weird, specific thing they wanted to make. That's the dream. Honestly, there's a real chance I can't pull it off. But the gap is right there, and somebody is going to fill it, and I'd love it to be us.

Who I am

I'm Scott Florida. I've been designing games in notebooks my whole life and never had the skills to actually build them, until now. You can reach me at [email protected].

About the name

"Project Daedalus" is a working codename, a nod to the mythological craftsman who actually built things that worked. It might not survive to launch (half the reason being that nobody's sure how to say it out loud, which is a real problem for a brand people are supposed to fall in love with), so the public name may change before this is a real product. The devlog is home either way.