I’m moving a data-driven feature from concept to production and need a supervised learning expert to make the core model work. The dataset is ready, but it still needs thoughtful cleaning, feature engineering, and the right algorithm choice so we can hit reliable accuracy in real-world use. Here’s what I want to achieve: you explore the data with me, recommend whether a classification or regression approach makes most sense, train and tune the model using mainstream Python tools—scikit-learn, TensorFlow, or PyTorch—then validate it with a clear metrics report. When the results look solid, we’ll export the trained model and wrap it in a lightweight REST API so my website team can drop it straight into our UI. Key deliverables • Cleaned, well-documented dataset • Reproducible training notebook or script • Trained model file with versioned weights • Evaluation report (confusion matrix, ROC/AUC or RMSE, plus brief interpretation) • Simple deployment guide or API stub for integration I’m happy to iterate quickly, share domain insights, and test candidate models as you progress. If you also feel comfortable advising on the AI-driven frontend experience, great—but the model itself is priority one.