senior-data-scientist

Run the experiment_designer.py script with an input data path and output path to structure the experiment. The references/experiment_design_frameworks.md guide covers step-by-step processes, architecture patterns, and troubleshooting for experiment setup.

Use the feature_engineering_pipeline.py script with --input and --output arguments to process datasets through a repeatable pipeline. The feature_engineering_patterns.md reference documents implementation examples, configuration practices, and deployment strategies.

The model_evaluation_suite.py script runs evaluation with a config file and produces structured JSON results. The skill documentation references Scikit-learn, XGBoost, MLflow, and Weights & Biases as the supporting evaluation and monitoring stack.

No, the scripts use only Python standard library modules such as argparse, json, logging, and datetime. They run with a standard Python 3 installation without installing additional dependencies.

The scripts provide scaffolding with validation and logging, but the core _execute methods contain placeholder logic returning success flags. You must implement the actual statistical computation, model training, and evaluation logic for your specific use case.