Build Databricks Apps faster with our brand-new Node.js + React SDK. Built for humans and AI.
AppKit is a TypeScript SDK for building production-ready Databricks applications with a plugin-based architecture. It provides opinionated defaults, built-in observability, and seamless integration with Databricks services.
AppKit simplifies building data applications on Databricks by providing:
- Plugin architecture: Modular design with built-in server and analytics plugins
- Type safety: End-to-end TypeScript with automatic query type generation
- Production-ready features: Built-in caching, telemetry, retry logic, and error handling
- Developer experience: Remote hot reload, file-based queries, optimized for AI-assisted development
- Databricks native: Seamless integration with SQL Warehouses, Unity Catalog, and other workspace resources
AppKit's power comes from its plugin system. Each plugin adds a focused capability to your app with minimal configuration.
- Analytics Plugin — Query your Lakehouse data directly from your app. Define SQL queries as files, execute them against Databricks SQL Warehouses, and get automatic caching, parameterization, and on-behalf-of user execution out of the box.
- Genie Plugin — Conversational AI interface powered by Databricks AI/BI Genie. Let users ask natural language questions against your data and get answers with automatic chart inference and visualization.
- Files Plugin — Browse, upload, and manage files in Unity Catalog Volumes. Supports multiple volumes, content type validation, and on-behalf-of user access.
- Lakebase Plugin — OLTP database operations against Databricks Lakebase with automatic OAuth token management. Returns a standard
pg.Poolcompatible with Prisma, Drizzle, TypeORM, and other ORMs.
Missing a plugin? Open an issue and tell us what you need — community input directly shapes the roadmap.
Follow the Getting Started guide to get started with AppKit.
🤖 For AI/code assistants, see the AI-assisted development guide.
📖 For full AppKit documentation, visit the AppKit Documentation website.
See CONTRIBUTING.md for development setup and contribution guidelines.