moai-domain-ml

ML data preprocessing pipelines transform raw tabular data into model-ready features through automated handling of numeric and categorical data, imputation, scaling, and one-hot encoding. This Skill provides modular components for reproducible feature engineering that integrate seamlessly with scikit-learn pipelines and modern ML frameworks, enabling end-to-end workflows from training through deployment.

Yes. This Skill builds robust preprocessing pipelines compatible with both PyTorch and TensorFlow frameworks. It supports production-grade MLOps integration, automated feature transformation, and monitoring across classification and regression tasks, enabling deployment-ready workflows at enterprise scale.

Feature engineering automation handles numeric and categorical transformations, missing-value imputation, and scaling reproducibly across large structured datasets. This Skill generates meaningful feature names post-transformation and integrates into modular preprocessing components that pair with experiment tracking and MLOps deployment tools.

Yes. This Skill integrates with AutoML platforms and experiment tracking systems as part of end-to-end ML pipelines. It provides reproducible preprocessing components that feed consistent, transformed features into model training, evaluation, and monitoring workflows.

Preprocessing reproducibility comes from modular, versioned feature transformation components that consistently apply imputation, scaling, and encoding rules. This Skill generates explicit feature names after transformation and maintains compatibility with scikit-learn pipelines, enabling identical preprocessing across different model types and deployment environments.