ai-ml-data-science
This skill structures the complete data science lifecycle—from problem framing and exploratory analysis through feature pipelines and model evaluation to production deployment. It emphasizes baselines first, leakage prevention, train-serve parity, and reproducibility using tools like LightGBM, scikit-learn, and Polars. Covers SQL transformation with SQLMesh, experiment tracking, drift monitoring, and operational handoff patterns.
ai-ml-data-science guides you through end-to-end ML workflows from problem framing and EDA to feature engineering, model selection, and production deployment.
AI-generated summary based on this skill's SKILL.md
Install
vasilyu1983/AI-Agents-public/ai-ml-data-science · repository language: Python
git clone https://github.com/vasilyu1983/AI-Agents-public
cp -r AI-Agents-public/frameworks/shared-skills/skills/ai-ml-data-science ~/.claude/skills/ai-ml-data-sciencenpx skillfed install vasilyu1983/AI-Agents-public/ai-ml-data-scienceFrequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
How do I build machine learning models from scratch with ai-ml-data-science?
ai-ml-data-science structures the complete workflow from problem framing through production deployment. Start with exploratory data analysis and feature engineering, establish baselines before complex models, then select and train models using tools like LightGBM or scikit-learn. The skill emphasizes leakage prevention, train-serve parity, and slice analysis during evaluation. Finally, document your model with contracts and lineage before deploying to production with MLOps pipelines.
What does ai-ml-data-science teach about preventing data leakage?
ai-ml-data-science prioritizes leakage prevention as a core design principle. It covers feature engineering practices that maintain train-serve parity, ensuring features computed at training time match exactly what's available at inference. The skill guides you through proper data splitting, temporal validation for time series, and feature store setup with tools like Feast to enforce consistency across training and serving environments.
Which tools does ai-ml-data-science recommend for model selection and evaluation?
ai-ml-data-science covers LightGBM and CatBoost for tabular data, scikit-learn for classical methods, and Polars for efficient data transformation. For evaluation, it teaches slice analysis, baseline comparison, and metrics selection. The skill also covers experiment tracking with MLflow and Weights & Biases, hyperparameter tuning with Optuna and Ray, and model interpretability using SHAP and LIME.
How does ai-ml-data-science address MLOps and production deployment?
ai-ml-data-science structures MLOps around continuous training pipelines, drift monitoring with tools like Evidently, and reproducible experiment setup. It covers data contracts, lineage tracking, and model card documentation for operational handoff. The skill includes production deployment checklists and patterns for monitoring model performance and data drift in live systems.
What SQL and data transformation practices are included in ai-ml-data-science?
ai-ml-data-science teaches SQL transformation workflows using SQLMesh for staging and feature computation. It emphasizes reproducible data pipelines that support both training and serving, with proper versioning and lineage tracking. The skill covers best practices for exploratory data analysis and feature engineering that maintain data quality and enable audit trails.
Does ai-ml-data-science cover time series forecasting and class imbalance?
Yes. ai-ml-data-science includes time series forecasting with LightGBM, addressing temporal validation and drift concerns specific to sequential data. For classification, it covers handling class imbalance through proper evaluation metrics, baseline strategies, and slice analysis. Both topics are integrated into the broader lifecycle framework emphasizing reproducibility and production readiness.
SKILL.md
rendered from the published skill — quoted content, verbatim
Data Science Engineering Suite - Quick Reference
This skill turns raw data and questions into validated, documented models ready for production:
- EDA workflows: Structured exploration with drift detection
- Feature engineering: Reproducible feature pipelines with leakage prevention and train/serve parity
- Model selection: Baselines first; strong tabular defaults; escalate complexity only when justified
- Evaluation & reporting: Slice analysis, uncertainty, model cards, production metrics
- SQL transformation: SQLMesh for staging/intermediate/marts layers
- MLOps: CI/CD, CT (continuous training), CM (continuous monitoring)
- Production patterns: Data contracts, lineage, feedback loops, streaming features
Modern emphasis (2026): Feature stores, automated retraining, drift monitoring (Evidently), train-serve parity, and agentic ML loops (plan -> execute -> evaluate -> improve). Tools: LightGBM, CatBoost, scikit-learn, PyTorch, Polars (lazy
(truncated - see the full file via the links below)
Read as markdown · JSON record · Browse the source repository
File tree — 15 files
frameworks/shared-skills/skills/ai-ml-data-science/SKILL.md
frameworks/shared-skills/skills/ai-ml-data-science/assets/eda/template-eda.md
frameworks/shared-skills/skills/ai-ml-data-science/assets/evaluation/template-evaluation-report.md
frameworks/shared-skills/skills/ai-ml-data-science/assets/evaluation/template-model-card.md
frameworks/shared-skills/skills/ai-ml-data-science/assets/features/template-feature-engineering.md
frameworks/shared-skills/skills/ai-ml-data-science/assets/project/template-quick.md
frameworks/shared-skills/skills/ai-ml-data-science/assets/project/template-standard.md
frameworks/shared-skills/skills/ai-ml-data-science/assets/review/experiment-review-template.md
frameworks/shared-skills/skills/ai-ml-data-science/data/sources.json
frameworks/shared-skills/skills/ai-ml-data-science/references/class-imbalance-patterns.md
frameworks/shared-skills/skills/ai-ml-data-science/references/data-contracts-lineage.md
frameworks/shared-skills/skills/ai-ml-data-science/references/eda-best-practices.md
frameworks/shared-skills/skills/ai-ml-data-science/references/evaluation-patterns.md
frameworks/shared-skills/skills/ai-ml-data-science/references/feature-engineering-patterns.md
frameworks/shared-skills/skills/ai-ml-data-science/references/feature-freshness-streaming.md