I operate a prize wheel that contains 50 distinct products, each assigned its own winning percentage and payout value. My goal is to analyse those probabilities, model every possible spin, and uncover the strategy that reliably drives maximum overall profitability rather than simply the highest hit-rate. Here’s the help I need: • Build or adapt an AI-powered model—feel free to use Python with NumPy/Pandas, TensorFlow, PyTorch, or any other framework you trust—that ingests the current wheel configuration (product list, individual odds, payout per product) and produces clear recommendations. • Run large-scale Monte-Carlo or similar simulations to estimate expected value, variance, and downside risk for each product and for the wheel as a whole. • Deliver a concise report and well-commented code notebook that highlight: – The optimal arrangement or spin strategy that maximises long-run profit. – Sensitivity analysis showing how profitability shifts if individual percentages or payouts are tweaked. – Any actionable insights (for example, removing low-margin products or re-weighting odds) backed by the model’s metrics. Acceptance criteria 1. Reproducible code that I can execute locally to rerun scenarios. 2. A results document (PDF or Markdown) explaining methodology, assumptions, and the recommended configuration. 3. Visualisations of projected profit distribution across 10,000+ simulated spins. If you thrive on probability theory, reinforcement learning, or optimisation problems and can turn raw percentages into clear, data-driven strategy, I’m eager to see how you’d tackle this project.