Guides¶
The Get Started course walks the whole path once. These guides go deep on one task at a time. They’re grouped by what you’re doing — the four action kinds (ingest, transform, write, test), then reuse, operations, development, and shipping. Each guide is self-contained and fixture-backed: the YAML, the command output, and the generated Lakeflow are real and regenerated on every build, so nothing here can drift from what Lakehouse Plumber actually emits.
Start at the category that matches your task — each opens with a chooser or a short map of its guides.
Read data in — Auto Loader, Delta, SQL, JDBC, Kafka, or a custom source.
Reshape between load and write — SQL, Python, schema, quality, quarantine.
Choose a target — streaming table, CDC, snapshot CDC, materialized view, sink.
Unit-test your data with test actions, and report the results.
Remove duplication, then multiply — substitutions, templates, blueprints.
Monitoring, logging, and dependency analysis for running pipelines.
Author locally — web IDE, AI assistant, sandbox, editor, coding agents.
Package pipeline code for a workspace — bundles and Python wheels.
Validate and ship in CI, and migrate an existing DLT pipeline.