AI Game Balance Simulator

You are a game balance engineer. Build an AI-driven game balance simulator: Game type: [INSERT - RPG, strategy, card game, shooter] Engine: [INSERT - custom simulation, Unity headless, Python] Balance scope: [INSERT - combat, economy, progression, all] Target playtime: [INSERT hours] Cover the full pipeline: 1. Balance model - define tunable parameters (damage, HP, costs, drop rates, XP curves), parameter space, constraints, and balance targets. 2. Simulation core - headless game runner, deterministic RNG, AI agents playing the game (heuristic, ML, or LLM-driven), parallel simulation. 3. Metrics and telemetry - win/loss rates, session length, resource flow, progression speed, build diversity, difficulty distribution, churn indicators. 4. Parameter sweep - grid search, Bayesian optimization, genetic algorithms, multi-objective optimization, sensitivity analysis. 5. Balance iteration loop - detect outliers (overpowered/underpowered), propose parameter changes, simulate, compare metrics, apply or reject. 6. Edge cases - extreme builds, min-max strategies, degenerate strategies, economy exploits, difficulty cliffs. 7. Human validation - playtest design, A/B tests, difficulty surveys, balancing against real player data.