I’m assembling a complete machine-learning pipeline and need someone who can take it from raw data all the way to a polished report. The data set will be a mix of structured tables and unstructured elements such as text or images, so your solution must gracefully handle both and tie everything together in Python. Core activities include data ingestion, cleaning, feature engineering, model selection and tuning. Evaluation must be demonstrated through both a train/test split and k-fold cross-validation, with clear error metrics that suit the problem type (classification or regression). I will rely on your plots—ROC curves, learning curves, residuals, and any other insightful graphs—as well as correlation heatmaps to explain feature relationships. An interactive dashboard (Plotly Dash, Streamlit, or a comparable Python framework) is expected so stakeholders can explore results themselves. Deliverables • Fully commented Python code or notebooks, ready to run end-to-end • Reproducible environment details (requirements.txt or environment.yml) • Technical report (PDF or Markdown) describing data prep, model logic, parameter choices, evaluation results, and key insights • Visual assets: high-resolution graphs & charts, correlation/heatmap figures, and the interactive dashboard (deployed or with clear startup instructions) Acceptance criteria will be met when the code executes without errors on my machine, metrics in the report match those produced by the script, and the dashboard renders the agreed visualizations interactively.