I’m working on a series of data-driven initiatives that aim to understand how customers behave across our digital touch-points, and I need an experienced machine-learning coder to jump in right away. The raw customer data is already flowing from our databases; what’s missing is the analytical horsepower to turn those records into clear behavioural insights that marketing, product, and support teams can act on. You’ll be cleaning and engineering features, running exploratory analyses, and building behaviour-focused models—think clustering, propensity scoring, or sequence analysis—using the tools you’re most comfortable with (Python/pandas, scikit-learn, TensorFlow, R, or a comparable stack). Along the way, I expect concise visualisations and well-commented code so every step is transparent and reproducible. Deliverables • A fully documented notebook or script that ingests the customer dataset, handles preprocessing, and trains at least one behaviour-prediction or segmentation model. • Exportable metrics, charts, or dashboards that illustrate key patterns you’ve surfaced. • A brief summary (1–2 pages) explaining methodology, findings, and next-step recommendations. Acceptance criteria • Code runs end-to-end on my environment with provided instructions. • Model performance and insight quality match or exceed baseline benchmarks we’ll agree upon at project kickoff. If you’ve tackled similar behaviour analysis projects before, I’d love to see a quick example or repo link when you reply.