What it does

This open-source VS Code extension wraps dbt Core (and Cloud) with IDE conveniences: autocomplete for models and macros, click-to-definition, query preview with CSV export and basic charting, column-level lineage, one-click runs of parent/child models and tests, and a live compiled-query preview as you type. It also bolts on some warehouse-specific extras like BigQuery cost estimation and SQL validation.

The interesting bit

The extension tries to solve the “dbt is just Jinja-flavored SQL in a text editor” problem by pushing project awareness down into the IDE—column lineage, health checks for missing columns or unmaterialized models, and even AI-generated documentation. The “defer to prod” feature is a nice touch for dbt Core users who want to skip rebuilding upstream models in dev.

Key highlights

• Auto-complete and go-to-definition for models, macros, sources, and docs
• Column-level lineage visualization (not just model-level)
• Generate dbt models from source definitions or convert raw SQL to dbt-ified code
• AI-assisted documentation generation with a UI editor that saves to YAML
• BigQuery cost estimator and SQL validator (catch typos, missing columns, bad parens)
• Force-tailed dbt logs viewer inside VS Code
• Works in dev containers, Codespaces, and WSL; requires dbt 1.0+

Caveats

• Several features (AI docs, health checks, cost estimation) likely lean on Altimate’s cloud service; the README doesn’t clarify what runs locally vs. what phones home
• The “AI” branding is vague—no mention of which models or whether API keys are needed

Verdict

Worth a look if you live in dbt + VS Code and want fewer round-trips to the docs site. Probably overkill if you just write simple models and already have a good dbt Cloud setup.