From the course: Orchestrating AI Agents with the Databricks Omnigent Meta-Harness
The role of the harness in multi-agent systems - Databricks Tutorial
From the course: Orchestrating AI Agents with the Databricks Omnigent Meta-Harness
The role of the harness in multi-agent systems
2026 is the year of the agentic harness, and 2027 will be the year of the meta-harness. What's an agentic harness? It's a tool that we use to give agents more power. Agents are an LLM plus tools. The harness adds context and memory management, as well as ways to interact with the world, such as MCPs and MCP connectors, as well as a prompt or a way to have an overarching identity for the agent. Meta-harnesses are ways that we can harness harnesses. And in this example, we're going to look at Omnigent, which is a popular open source meta-harness from Databricks. And we're gonna use Omnigent to have Codex and Pi help us build an application and then create our own custom agent on top of it. Let's get started. you
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.
