Lakebase Cookbook

Examples and guides to accelerate your Databricks Lakebase projects.

Lakebase is Databricks’ fully-managed, Postgres-compatible OLTP database, built for transactional workloads that sit alongside your lakehouse.

Examples in this cookbook

Example Description
GraphRAG Knowledge-graph-augmented RAG with pgvector seeding and recursive-CTE graph traversal
Branching CI/CD Validate schema changes on an isolated Lakebase branch in GitHub CI/CD, with an AI impact report per PR
Consort Spec-first, test-driven agentic development where every green is a real test on a live Lakebase branch, held by a deterministic state machine and human gates
SCM Extension A VS Code / Cursor extension that pairs each code branch with a Lakebase database branch
SCM Utils The portable engine behind the extension and Consort: branching, the paired-branch SCM state machine, credentials, and migrations
AI Memory Short-term, long-term, and semantic (pgvector) agent memory on one Lakebase store
FastAPI App A Databricks App serving a Lakebase synced table through a FastAPI REST API
Genie Caching A pgvector semantic cache in front of Genie Spaces, so reworded questions reuse SQL instead of regenerating it
Feature Store Feature store integration
Reverse ETL Reverse ETL patterns

Contributing

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes
  4. Submit a pull request

Guidelines:

  • Resources related to your example must be deployable via Databricks Asset Bundles
  • Keep database configurations small to reduce spend
  • Use uv as the package manager
  • Use ruff as the linter
Edit this page →