Databricks Showcases Expanded Genie Agent Capabilities for Deeper Analysis
This article details several enhancements to Databricks Genie Agents, including Agent mode APIs for multi-step reasoning, support for analyzing unstructured data from Unity Catalog volumes, and improvements to Genie Code for agent curation. These updates enable more complex business questions, allow agents to combine structured and unstructured data insights, and streamline custom agent development. The expanded capabilities aim to make Genie Agents more powerful and easier to create, benefiting data analysts, developers, and business users leveraging AI agents. Genie Agents were initially announced at Data and AI Summit, and these updates build on that foundation.
- →Agent Mode APIs Enable Multi-Step Analysis for Databricks Genie Agents
- →Genie Agents Analyze Unstructured Data from Unity Catalog Volumes
- →Genie Code Enhances Curation and Management of Genie Agents
- →Share Genie Agent Conversations for Collaboration and Feedback
Notes (4) ›
- Agent Mode APIs Enable Multi-Step Analysis for Databricks Genie Agents
Agent mode provides an agentic reasoning loop for multi-step analysis, enabling Genie Agents to create research plans and return reports for complex business questions. Developers can integrate this capability into custom applications and tools using new Agent mode APIs, which support streaming responses via Server-Sent Events (SSEs) and include visualization support.
- Genie Agents Analyze Unstructured Data from Unity Catalog Volumes
Genie Agents can now analyze unstructured files such as PDFs, documents, slide decks, and images stored in Unity Catalog volumes, combining insights from these with structured data. This capability respects Unity Catalog permissions, ensures data governance, and utilizes content search indexing for faster, higher-quality retrieval across large volumes.
- Genie Code Enhances Curation and Management of Genie Agents
Genie Code offers enhanced tools for building, diagnosing, and managing Genie Agents, allowing authors to describe an agent's purpose and data sources to create a high-quality baseline. It can also analyze conversation failures or benchmark runs to propose improvements and help manage agents in production by summarizing feedback trends and suggesting context adjustments.
- Share Genie Agent Conversations for Collaboration and Feedback
Users can now share chats from specific Genie Agents with teammates and agent authors, with shared conversations remaining up-to-date with new messages and analyses. This feature supports insight democratization and allows for reviewing conversation quality and responding to feedback.
https://www.databricks.com/blog/expanding-genie-agents-deep-analysis-file-reasoning-and-more
Related releases
- Databricks Unity Gateway API now supports programmatic resource management Databricks Release Notes ·
- Schema-level HTTP connections in Databricks Unity Catalog are now GA Databricks Release Notes ·
- Databricks-managed MCP Connectors Integrate with Unity Gateway in Beta Databricks Release Notes ·
- Databricks-managed MCP Connectors Integrated with Unity Gateway (Beta) Databricks Release Notes ·
- Databricks Unity Gateway Achieves General Availability for Enterprise AI Governance Databricks Release Notes ·
- Unity Gateway API: Manage resources with SDKs, CLI, and Terraform Databricks Release Notes ·