I need a complete, production-ready machine-learning pipeline that solves a classification problem. The core outcome is a trained, well-evaluated model along with clean, reproducible code and concise documentation that explains each step—from data preparation through deployment-ready export. You’ll advise me on the most suitable algorithms, features, and evaluation metrics, then iterate until we hit an agreed-upon performance target. I can supply data if needed, but I’m equally open to you recommending sources or strategies for gathering additional training samples should that prove advantageous. Please use Python with familiar libraries such as scikit-learn, TensorFlow, PyTorch, or similar, and deliver: • Notebook or scripts that ingest the data, perform preprocessing, train, validate, and test the model • Clear explanation of hyper-parameter choices and feature engineering decisions • Exported model artifact plus a lightweight inference script or API example • A brief read-me covering setup and replication steps If you’ve previously shipped classifiers in real-world settings, tell me which techniques and tools you favour and why. I’m ready to get started right away and will provide prompt feedback on each milestone.