Databricks has unveiled the Public Preview of Databricks Apps, a significant advancement aimed at simplifying the process of building and deploying internal data and AI applications on major cloud platforms such as AWS and Azure. This launch resonates with the growing emphasis on streamlined development workflows in an environment where data-driven decisions are paramount.
The introduction of Databricks Apps offers developers a powerful toolkit for creating applications focused on data visualization, AI deployment, self-service analytics, and data quality monitoring. A notable feature is its capability to invoke specific AI models, granting developers greater flexibility while maintaining a robust governance framework through tools like Unity Catalog. This feature enables developers to manage data compliance efficiently and establish secure user authentication with OpenID Connect (OIDC), OAuth 2.0, and Single Sign-On (SSO).
Databricks Apps supports a variety of popular frameworks, including Dash, Shiny, Gradio, Streamlit, and Flask. This diverse support allows developers to design applications that are user-friendly and accessible even to those without a technical background. For instance, companies can create custom dashboards to showcase key performance metrics, making it easier for non-technical team members to interpret complex data. A prime example includes leveraging sentiment analysis on customer feedback or building predictive models to enhance sales forecasting accuracy.
One of the standout advantages of Databricks Apps is its integrated environment for application deployment. Once constructed, these applications can be launched and managed directly within the Databricks platform, significantly reducing the burden of additional infrastructure management. The Unity Catalog further ensures that data access controls are adhered to within the applications, aligning with a unified governance model to promote data integrity and security.
The ability to deploy applications using serverless computing is another feature that distinguishes Databricks Apps. This approach alleviates the need for developers to set up separate infrastructure for their applications. As a result, teams can focus more on innovation and less on configuration. Developers also have the flexibility to work in their preferred IDEs, such as Visual Studio Code and PyCharm, while leveraging tools for version control and continuous integration/continuous deployment (CI/CD) pipelines.
Upon the creation of applications, a unique URL is generated, facilitating easy sharing among users. This promotes collaboration and encourages exploration of internal applications, recently highlighted by Lukas Heidegger from E.ON Digital Technology. Lukas noted that integrating Databricks Apps into their DevOps processes allowed for rapid feature demonstrations and testing, all while providing a secure, production-ready front end for their internal applications—achieved without additional infrastructure concerns.
Security remains a top priority with Databricks Apps. Sensitive data stays within the Databricks environment unless shared externally, and various security measures—such as granular access controls and automatic user authentication—are in place. Moreover, the Unity Catalog adds lineage capabilities, enabling organizations to track data usage and better comply with regulatory requirements.
Heather Gomer from SAE International shared her experience, stating, “Databricks Apps helped me turn my RAG proof of concept into a polished and branded application. We built a RAG system to answer user questions by utilizing our company’s extensive knowledge base.” This illustrates how Databricks Apps can assist in transforming initial ideas into fully functional applications that meet organizational needs.
The range of applications that can be built with Databricks Apps is extensive. Companies can design customized data visualizations, create AI applications for predictive analysis, develop self-service analytics tools, and set up tools for data quality monitoring. Lukas further explained that their organization fully realized its user-facing data interfaces through Databricks Apps, emphasizing its efficiency in hosting dashboards featuring advanced functionalities like semantic search tools.
Industry leaders have recognized the value of this new offering. Tareef Kawaf, CEO of Posit, highlighted the importance of code-first tools for enabling organizations to extract meaningful insights from their data, while Dave Gibbon from Plotly expressed excitement about the opportunities Databricks Apps presents for enhancing analytics delivery to business users.
In conclusion, the launch of Databricks Apps represents a pivotal step in the evolution of AI and data applications. By focusing on simplifying the development process, enhancing security measures, and promoting user-friendliness, Databricks is set to empower organizations to harness their data more effectively and drive informed decision-making.
