@Suheb, Depends on your usecase.

However, if it fits, I would recommend that you start with a multi-agent supervisor if you have the agents from the list below

• An existing Agent Bricks: Knowledge Assistant(/generative-ai/agent-bricks/knowledge-assistant.md) agent endpoint.
• An existing Genie space. To set up a Genie space, see Set up and manage an AI/BI Genie space.
• AI agent tools created as Unity Catalog functions. See Create AI agent tools using Unity Catalog functions.
• External MCP servers with configured Unity Catalog connections. You can install an MCP server from Databricks Marketplace or Use external MCP servers. The connection must use bearer token authentication or OAuth Machine-to-Machine authentication. See Authentication methods for external services.

You may also use Mosaic AI agent framework using Langgraph. Something similar to this - https://docs.databricks.com/aws/en/generative-ai/agent-framework/multi-agent-genie