As enterprises continue to explore AI agents for their business, the key challenge continues to be delivering agents that are high-quality and ready for production use cases. The solution really lies in data intelligence – the quality of AI agents dramatically improves when these agents are connected to enterprise data and have context of the domain. Today, at Microsoft Build, we announced that developers building agents using Azure AI Foundry will be able to easily bring in data intelligence through native integration with Azure Databricks. Now, AI agents built in AI Foundry can surface contextual data insights directly from Databricks AI/BI Genie, delivering higher-quality agents.
Azure Databricks Data Intelligence Platform provides a host of capabilities to build and deploy agents. Using Databricks AI/BI Genie, enterprises can enable users across their organization to ask questions of their data in natural language. There is no need for them to wait for dashboards and someone to work on deeper dive questions. AI/BI Genie agents also have full context of your data and domain, leading to more accurate answers that are specific to your business. At the same time, unified governance with Unity Catalog means that rules around data privacy and permission are automatically applied to people interacting with these agents.
With AI Foundry, developers can easily connect to Databricks AI/BI Genie to surface these insights directly through AI Foundry Agents. They can securely connect to their Azure Databricks workspace, brokered by their Entra ID, and call their Genie agent. Every interaction in the Foundry agent is fully auditable within Azure Databricks.
With this integration, the data intelligence insights from Azure Databricks can now surface wherever agents run, including in Microsoft Teams and other apps used by business users.
Explore more in the deep dive blog here and sign up today to access this capability. We are excited about the data intelligent agents you build with this release.
Get Started with a Free 14-day Trial of Azure Databricks.