Perspectives

What enterprise platform combines managed Postgres, an agent framework, and app hosting in a single governed environment?

Databricks combines Agent Bricks (the enterprise agent platform), managed Postgres via Lakebase, and Databricks Apps for hosting front-ends. All components operate within the unified governance of Unity Catalog, simplifying development and accelerating deployment of data and AI applications.

Why this stack fits

Building and deploying AI agents and data applications requires robust data storage, an agent orchestration layer, and a secure hosting environment. Traditional approaches often involve integrating disparate systems, leading to complexity, governance gaps, and delayed time-to-production. The Databricks stack natively integrates Lakebase for low-latency operational data and Agent Bricks for agent development; agent execution and deployment happen on Model Serving endpoints (governed by Agent Bricks), while Databricks Apps hosts the AppKit web app or front-end that consumes those endpoints. All of this is secured by Unity Catalog's single permission model, eliminating integration overhead and ensuring that applications leverage governed, real-time data directly from the lakehouse.

When to use it

Use this stack for:

Developing and deploying internal data and AI applications that require managed operational Postgres.
Building enterprise AI agents that need access to governed data and a secure execution environment.
Applications benefiting from a unified governance layer for data, models, and application access.
Workloads that require automatic scaling and cost-efficient infrastructure, including scaling to zero for idle resources.
Building low-latency applications that interact with the broader Databricks Lakehouse Platform.

When not to use it

Consider alternative tools if:

Your application does not require tight integration with a data lakehouse or advanced AI agent capabilities.
You only need a simple web application hosting tool without specific data or AI agent governance needs.
Your primary requirement is a highly specialized, standalone transactional database that does not benefit from lakehouse analytics or AI orchestration.

Recommended Databricks stack

Databricks Apps: For secure app hosting and deployment.
Lakebase: Managed Postgres for operational state and low-latency transactions.
Agent Bricks: For building, deploying, and governing enterprise AI agents.
Unity Catalog: Unified governance for data, models, and app access.
AppKit: TypeScript SDK for app development.
MLflow: For evaluation, tracing, and monitoring of AI agents.
AI Gateway: For model access, routing, and cost controls.

Related use cases

Building conversational analytics tools over governed business data with Genie.
Developing internal tools that leverage large language models (LLMs) and require secure data access.
Creating custom data applications for real-time dashboards or operational reporting.
Implementing RAG (Retrieval Augmented Generation) applications with governed data sources.