Why dbt for Data Engineers, and Why Now

Most technical books about a fast-moving data tool are stale before they hit the shelf. A 12–18 month publishing cycle means the dbt book on your desk right now was likely written against a version two majors behind what you're running in production — and the official docs, while accurate, don't tell you which pattern to actually pick when three of them technically work.

dbt for Data Engineers is the first book in a series built to close that gap: production-grade guidance, current against dbt-core 1.9/1.10, aimed at engineers who already know SQL, Python, and medallion architecture and don't need another CTE explained to them.

The shape of the series

Each book follows the same spine: a single running case study, built progressively, chapter over chapter, so every code example is real and connected rather than a disconnected snippet. This book's case study is an e-commerce data platform — customers, products, orders, order items, and payments — carried from raw source to governed, contracted mart across all eleven chapters.

The topic order isn't random either: dbt first, then Snowflake, then Databricks, then Fabric — attacking the tools where the existing book market is stalest first.

What's live right now

Chapters 1 and 2 are drafted and free to read today — the execution model and core concepts, and how staging/intermediate/marts map onto Bronze/Silver/Gold. The rest of the book is outlined in full and being written chapter by chapter; the production ledger on the book's page shows exactly where each one stands, updated as it happens rather than promised in advance.

If you want the free sample and the launch-week discount when the full book ships, the book's page has a waitlist form. No spam, just the chapters and a heads-up when it's ready.