This Might Be the Nerdiest Thing I’ve Ever Done
I designed a board game with zero dice and zero luck.
Yes, I know how that sounds. But hear me out.
The combat system uses blind bidding—both players secretly commit resource cards, then reveal them simultaneously. The supply lines actually matter: if your troops can’t trace a path back to one of your cities, they disband. There are technology trees, diplomatic backstabbing mechanics (tear up a treaty and your stability plummets), and seven asymmetric factions based on the Warring States period of ancient China.
I’ve playtested it with strangers. Their feedback was consistent: “Great game. Exhausting to resolve.”
Turns out when every battle involves secret commitments and response chains, the mental overhead of running the rules yourself is brutal.
So I’m doing the rational thing: building a computer vision system that does it for you.
Here’s the idea: a camera mounted above the table watches where every piece moves. A rules engine I’ve already designed (think: state machine that handles LIFO response stacks, hidden information synchronization, and modular rule protocols) resolves everything automatically. A speaker announces the results. Players just play.
No screens. No apps. No tapping virtual buttons. You move physical pieces on a physical board. The system watches, listens, and speaks only when it needs to. The rest of the time, it shuts up and lets you stare at your opponent across the table.
What I’ve done
I’ve spent years designing the modular rule engine (V12.4) and two complete games on top of it. Both have been blind-playtested with real feedback. The rule documentation is written—not in board game jargon, but in logic trees that map directly to code.
What I can’t do
The computer vision piece. The real-time state machine that actually runs the rules instead of just describing them on paper.
I need someone who can.
What I’m looking for
Someone who looks at a problem like “track 20+ physical game pieces in real time from a fixed overhead camera” and thinks “that’s a fun Saturday”—not “where’s the nearest API that already does this.”
Someone who understands that a LIFO response chain isn’t a bug in the rules, it’s a feature, and building a state machine that handles it correctly is the whole point.
This project sits at the intersection of computer vision, state machine design, and real-time systems. It’s not another CRUD app. It’s making a computer understand what’s happening on a physical table.
What I’m offering
A product that already exists. Not an idea. Not a pitch deck. Physical prototypes. Playtested. People have played these games and said “I’d buy this if the resolution wasn’t so heavy.”
A technical specification disguised as a rulebook. Every mechanic is documented precisely enough that you can trace each line of code back to a specific rule.
A 4-month MVP vesting agreement. You build the vision module and the real-time rules engine. I handle product, market, and fundraising. Deliverables met = 25% technical equity. In writing.
A project that won’t make you cringe when you describe it to your engineer friends at parties.
The ask
If you’re in Europe, the UK, or anywhere reachable by email and a video call—let’s talk. I’ll show you the prototype over a screen share, walk you through the rule architecture, and you can decide if this is the kind of problem you want to solve.
My email: caomark@yeah.net
If you’re not the right person but know someone who might be—someone who works in computer vision, builds game engines, or just really likes state machines—please forward this to them.