Over Christmas break, I built GLITCH - a first-person survival shooter where you play as a "glitch" fighting debugging agents in a neon-lit spherical arena.
The stack:
React Three Fiber for 3D rendering
Zustand for state management
PostgreSQL for the global leaderboard
Hosted on Replit
What I built in ~40 prompt iterations:
- Full 6-DOF movement (fly in any direction)
- 4 enemy types with unique AI behaviors (rushers, floaters, brutes, glitchers that teleport and drop mines). If it's interesting, I have plans to expand.
- Shield ability with knockback physics
- Wave-based progression system
- Global leaderboard with nickname entry for top 10 scores
- Prodigy-style electronic soundtrack with random track rotation
- Visual effects: chromatic aberration on damage, scanlines, vignette
Interesting challenges we solved with AI assistance:
- Elastic boundary system to prevent players getting stuck at arena edges
- 3-meter knockback limit so enemies get pushed back but resume chasing
- Mine cleanup system with proper TTL handling
- The entire game loop - from concept to playable with leaderboards - was built conversationally with an AI coding assistant. Each prompt built on the last, iterating on mechanics until they felt right.
I actually wanted to build something I like to play instead of "consuming" others' games.
note: I'm passionate about this too. We're building https://fairlight.app/ as a successor to https://moneyflow.camplight.net/ (an internal tool that helps accounting and book-keeping we use to power our venture studio and agency as it's a real hell)
Imho, The long term goals may be similar to what "normal" companies may have. Cooperatives are just means to an end but generally one may achieve lofty ambitions in any type of organization. The document sums up very nicely the pros and cons of cooperative targeting freelancers. The only thing one has to decide is wether one will reach sustainability in a normal company or in a cooperative - the journeys are two different wild adventures.