How do data analysts build and ship their own apps?

Summary

• Data analysts build and ship their own apps with Databricks Apps, which lets you build and deploy secure data and AI applications directly on Databricks with no separate infrastructure to manage. See Databricks Apps.
• Use familiar Python frameworks: build with Streamlit, Dash, Gradio, Flask, or Shiny (and Node.js frameworks such as React), start from pre-built templates, and iterate quickly in your preferred editor.
• Apps run on serverless compute that Databricks provisions automatically — no clusters, containers, or scaling to manage. Apps are billed per hour of compute while running.
• Governed data access is built in: apps read and write Unity Catalog data and query Databricks SQL warehouses using the app's identity, so analysts don't manage connection strings, and data lineage stays visible.
• Ship in a few steps: create an app from + New, write code, and deploy to get a shareable URL — with built-in SSO/OAuth authentication and support for Git and CI/CD via Declarative Automation Bundles. See Get started with Databricks Apps.

How do data analysts build and ship their own apps?

Data analysts often have the insight and the data but not a place to turn it into an interactive tool for their team. Databricks Apps closes that gap: it lets you build and deploy secure data and AI applications directly on the Databricks platform, with no separate infrastructure to stand up. Apps run as containerized services on serverless compute and integrate with Unity Catalog for governance, Databricks SQL for querying, and OAuth for authentication — so an analyst can go from a Python script to a shared, governed app. See Databricks Apps.

Why Databricks Apps for data analysts

• Build with frameworks you already know. Databricks Apps takes a bring-your-own-framework approach: build with Python frameworks like Streamlit, Dash, Gradio, Flask, and Shiny, or Node.js frameworks such as React. Pre-built templates get you started fast so you can build interactive dashboards, data-entry forms, RAG chat apps, and operational tools with code rather than standing up a web stack. See Introducing Databricks Apps.
• No infrastructure to manage. Apps run on serverless compute that Databricks provisions automatically — it handles hosting, scaling, and maintenance, so you focus on the app, not the servers. Apps are billed per hour of compute time while running.
• Governed data access out of the box. Apps access workspace resources like SQL warehouses and account resources like Unity Catalog, each with its own configuration, identity, and isolated runtime. Apps use their service principal credentials to read and write tables without manual connection-string management, and Unity Catalog lineage keeps data origins and usage visible for traceability and compliance.
• Authentication is built in. Apps include out-of-the-box SSO and authentication using OIDC/OAuth 2.0, enforcing user-level access based on Unity Catalog permissions. Users sign in through workspace SSO, so there is no separate login system to build.
• Ship and share quickly. Create an app from + New, make changes in your favorite editor, and deploy. Each deployed app gets a unique URL you can share, and colleagues can discover apps from the Compute tab. See the Databricks Apps documentation.
• Grow into production practices. Apps integrate with Git version control and CI/CD, and you can automate deployments and promote across dev/stage/prod using Declarative Automation Bundles. See the bundles apps tutorial.

Getting started

• Follow Get started with Databricks Apps to build and deploy your first app.
• Read Introducing Databricks Apps and the general availability announcement.
• Review the Databricks Apps best practices before you ship.

FAQs

Do I need frontend or DevOps skills to build a Databricks App?

No. You build with familiar Python frameworks like Streamlit, Dash, or Gradio (or Node.js frameworks), starting from pre-built templates, and Databricks handles hosting, scaling, and maintenance on serverless compute.

How does a Databricks App access my data securely?

Apps connect to Unity Catalog data and Databricks SQL warehouses using the app's own identity and credentials, so access follows Unity Catalog permissions and lineage is tracked — no manual connection strings required.

How do I deploy and share an app?

Create an app from + New, write your code, and deploy. Each app gets a unique URL to share, protected by built-in SSO, and colleagues can find it from the Compute tab.

Can I use Git and CI/CD with Databricks Apps?

Yes. Apps integrate with Git version control and CI/CD pipelines, and you can automate deployments and promote across dev/stage/prod using Declarative Automation Bundles.