I need a hands-on data scientist to design and implement a predictive analytics pipeline that segments our customers into meaningful groups. My raw material is customer data—demographics, behavioural events, transaction history—already available in CSV and easily exported from our CRM. Here’s what success looks like for me: • A clear data-cleaning and feature-engineering workflow in Python (Pandas, NumPy, scikit-learn preferred). • At least two clustering approaches (e.g., K-means and hierarchical) benchmarked with appropriate metrics such as silhouette score. • A concise Jupyter notebook explaining the methodology, hyper-parameter choices, and visualisations that make each segment easy to interpret for marketing teams. • A short summary slide deck highlighting the segments’ defining attributes and actionable recommendations. I can provide sample datasets immediately and will be available for quick feedback cycles. Once the notebook runs end-to-end on my machine and the insights are clear enough to guide targeted campaigns, the job is done.