Concepts

Read these pages in order — each builds on the previous:

1. Artifacts & the DAG — The mental model: files drive the graph
2. Dependencies — Declaring what a stage reads
3. Outputs — Declaring what a stage produces
4. Loaders — How files are read and written
5. Parameters — Typed configuration for stages
6. Pipelines — Registration, composition, and discovery
7. Fingerprinting — How Pivot detects code changes
8. Caching & Skip Detection — When stages skip and how cache works