databricks-model-serving

Deploy MLflow models to Databricks model serving endpoints by registering trained models and creating scalable serving configurations. The Skill supports traditional ML models, custom PyFunc models, and GenAI agents for direct endpoint querying.

Yes, you can deploy GenAI agents including LangGraph and ResponsesAgent to Databricks model serving endpoints. These deployed agents can integrate directly with Unity Catalog functions or Vector Search tools for enhanced querying capabilities.

Check the status of a Databricks model serving endpoint using the Databricks SDK or MCP tools. These interfaces allow you to query endpoints, verify deployment progress, and route traffic to manage your model serving infrastructure effectively.

Databricks model serving supports integration with Unity Catalog functions and Vector Search tools. This allows deployed GenAI agents and MLflow models to leverage centralized data governance and vector similarity search within your endpoint queries.

Manage traffic routing for Databricks model serving endpoints using the Databricks SDK or MCP tools. This approach enables you to route traffic, verify endpoint status, and query deployed MLflow models and GenAI agents across scalable configurations.