Where the Databricks world meets.

How Enterprises Govern AI Agents Across Multiple Models

Databricks announced the general availability of the Unity AI gateway to provide centralized multi-model governance, cost controls, and end-to-end observability for enterprise AI agents. Panelists discussed how coding agents and harnesses are evolving beyond programming into long-running operations, personal software development, and automated organizational workflows.

Demo: Building a Governed AI Agent with Unity AI Gateway

This video demonstrates how to build, update, and govern a store operations AI agent using Databricks Agent Bricks and the Unity AI Gateway. The tutorial highlights integrating custom Model Context Protocol servers, recording execution traces with MLflow, and enforcing security policies and budget controls.

Stellantis: How a Global Automaker Deployed AI Agents at Scale

The Ontological Definition of Databricks Genie Ontology via OntoRank

Databricks Genie Ontology uses a ranking algorithm (OntoRank) to dynamically infer data definitions based on source credibility, usage frequency, and certification, enabling AI agents to answer questions without requiring formal ontologies. The tool democratizes data access but risks surfacing incorrect definitions if not paired with basic data governance practices.

DABs: UI to DABS sync

Building Governed Agents with Databricks

Databricks extends its Unity Catalog governance layer to AI agents, MCP servers, and tools to solve "agent sprawl" by providing centralized discovery, access controls, and audit logging. The system allows developers to quickly build agents while IT teams enforce fine-grained policies, manage credentials, and track end-to-end request and response lineage.

From Primitives to Production: How Anthropic Builds Agents

Anthropic defines agents as LLMs in loops with tool access, leaning on model intelligence over rigid workflows, and uses domain-specific skills, MCPs, and evals to build effective agents across sectors. A demo showed a site reliability agent autonomously identifying a database pool issue, fixing the code, and generating a postmortem.

DABs: watch flag

DABs: Let policies rule your cluster