Back to all cases
CASE STUDY / AI & Data Science
Slot Game Engine & Probability Optimizer
Digital Gaming, Probability Design & Simulation3×3 slot game engine combining Monte Carlo simulation with heuristic search to auto-optimize reel configurations under RTP 95% and win rate ≥ 55% constraints. Full-stack implementation with FastAPI backend and React frontend.
Delivery Details
This is a slot game engine and probability optimization system. The core challenge is designing reel configurations that simultaneously meet RTP (Return to Player) and win rate targets. The system uses heuristic search iterating 800 steps, evaluating each configuration via 50,000-spin Monte Carlo simulation, with final 100,000-spin verification. Five symbols (varying multipliers) combined with five winning patterns (horizontal, vertical, diagonal, V-shape, custom), using symbol distribution mutation and adaptive search to efficiently converge in a massive search space. Full-stack implementation supports both CLI and API modes, with a React frontend for real-time gameplay and statistics.
Industry Context
Probability-based games must balance mathematical return, win rate, volatility, and player experience, while any reel or payout change can alter the full outcome distribution. Large-scale simulation, reproducible randomness, and multi-objective search are essential for validating designs against constraints.
Business Problem
Game probability design had to balance RTP, win rate, and player experience under costly manual tuning.
What I Built
Used Monte Carlo simulation and heuristic search to evaluate large reel-configuration spaces.
Business Impact
Reduced multi-day manual tuning into minute-level optimization with reproducible validation.
For career conversations, advisory work, or a concrete Data & AI project, feel free to get in touch.