I want to turn my manual forex and crypto playbook into fully automated, AI-enhanced systems. The core logic covers every style I currently trade—trend following, mean reversion and arbitrage—so the code needs to be modular enough to let me switch any of those on or off, or run them simultaneously. Python is my language of choice because of its rich ecosystem (pandas, NumPy, scikit-learn, TensorFlow, MetaTrader-Python, CCXT, etc.), and I’d like the final solution to sit cleanly on that stack. Machine-learning elements should focus on signal generation and dynamic risk management, not just curve-fitting; think rolling feature engineering, online learning or reinforcement learning methods where appropriate. Integration is flexible. If you have a strong preference for MetaTrader, TradingView or a direct Binance API connection, let me know which route best matches the architecture you plan to build—latency, order management and data fidelity will ultimately drive the decision. Deliverables • Well-structured Python project (clear separation of data, strategy logic, ML modules, execution) • Back-testing suite with walk-forward and Monte Carlo analysis • Live-trading bridge that can run on a VPS or cloud container • Read-me style documentation plus a short video walkthrough so I can expand modules myself I’ll provide exact strategy parameters, broker endpoints and historical data once we kick off. Let’s build something robust, transparent and easy to extend.