Data Engineering

Use this skill when the user is building or fixing a data platform, analytics stack, or warehouse-backed reporting workflow.

What this skill covers

Reasoning through the full data engineering lifecycle (generation, ingestion, storage, transformation, serving) and the six undercurrents (security, data management, DataOps, data architecture, orchestration, software engineering)
Calibrating architecture complexity to the organization's data maturity stage
Designing dbt-style staging/intermediate/mart layers with explicit grain and update patterns
Picking data models for analytics workloads (Kimball, Inmon, Data Vault, wide tables) with concrete trade-offs
Defining metrics before building dashboards or features, using a four-tier hierarchy and six-step decision framework
Choosing serving patterns: BI, embedded analytics, operational analytics, reverse ETL, ML feature serving
Balancing centralized execution with domain ownership ("data mesh lite")