Optimize data features for machine learning models.
Feature Engineering Strategist
Feast, Tecton, SHAP, feature pipelines
Best for
- ▸Designing time-based aggregation features (7d, 30d, 90d windows) for fraud detection models
- ▸Building feature stores with Feast or Tecton for production ML serving at scale
- ▸Creating target encoding pipelines with proper cross-validation to prevent leakage
- ▸Implementing real-time feature serving architectures for sub-100ms prediction latency
What you'll get
- ▸Complete feature engineering pipeline with pandas/polars code, including categorical encodings, temporal aggregations, and cross-validation strategies
- ▸Feature store architecture diagram with Feast configuration files, feature definitions, and serving layer specifications
- ▸Feature importance analysis with SHAP explanations, feature selection recommendations, and performance impact metrics
ML problem context (classification/regression/ranking), available raw data sources, serving requirements (batch vs real-time), and any domain constraints or business logic.
Detailed feature engineering strategy with specific transformation code, feature store architecture designs, encoding recommendations, and production serving patterns.
What's inside
“You are a Feature Engineering Strategist. You transform raw data into production-ready ML features by combining statistical expertise, domain knowledge, and systematic methodology to prevent leakage, training-serving skew, and feature decay. - **Domain-Driven Over Automated**: Create 20 thoughtful f...”
Covers
Not designed for ↓
- ×Raw data collection or ETL pipeline design (that's data engineering)
- ×Model architecture selection or hyperparameter tuning (that's ML engineering)
- ×Basic data cleaning or exploratory data analysis workflows
SupaScore
87.53▼
Evidence Policy
Standard: no explicit evidence policy.
Research Foundation: 7 sources (3 paper, 3 official docs, 1 books)
This skill was developed through independent research and synthesis. SupaSkills is not affiliated with or endorsed by any cited author or organisation.
Version History
v5.5 final distill
Pipeline v4: rebuilt with 3 helper skills
Initial release
Prerequisites
Use these skills first for best results.
Works well with
Need more depth?
Specialist skills that go deeper in areas this skill touches.
Common Workflows
ML Model Development Pipeline
End-to-end ML pipeline from feature design through experimentation to production deployment
© 2026 Kill The Dragon GmbH. This skill and its system prompt are protected by copyright. Unauthorised redistribution is prohibited. Terms of Service · Legal Notice