All videos
0:00 / 0:00
ai

Introducing Omnigent: an open meta-harness – Matei Zaharia, Co-founder and CTO, Databricks

Databricks24 June 2026Watch on YouTube

Part of series

Ep. 2 · Omnigent: AI Agent Orkestratie

Databricks introduceert Omnigent, het open-source meta-framework dat meerdere AI-agenten combineert in één gecontroleerde, interoperabele workflow.

View the series

What you'll learn

  • Omnigent is an open-source meta-harness that makes AI agents interoperable, allowing them to work independently of the underlying model or platform
  • An AI agent harness is a layer in the agentic stack that manages sessions, policies and skills separately from the agent implementation
  • Too many different harnesses create fragmentation, while Omnigent provides standardization so skills and policies are reusable

Frequently asked questions

What is the main problem that Omnigent solves in AI-agent infrastructure?
Omnigent solves the interoperability problem: without standardization, AI agents, their sessions and skills are locked to one harness or model, preventing users from switching platforms and making skills non-reusable.
How does Omnigent differ from existing AI-agent harnesses?
Omnigent is a meta-harness that sits above existing harnesses and agents, keeping sessions, policies and skills independent and shareable without being locked to a single implementation.
Why did Databricks choose to release Omnigent as open source?
Open-source Omnigent promotes broad adoption, industry standardization and collaboration around AI-agent interoperability, allowing the entire ecosystem to benefit.

Topics

ai agentsinteroperabilityopen sourceagent infrastructuredatabricks

Description from the channel

Introducing Omnigent: A Meta-Harness for AI Agents Omnigent makes AI agents interoperable, so sessions, policies, and skills stay with users regardless of where the agent or model is running. Learn more about Omnigent: https://www.databricks.com/blog/introducing-omnigent-meta-harness-combine-control-and-share-your-agents 00:00 — Why the agentic stack needs a new layer 01:55 — What's an AI agent harness? 02:40 — The problem with too many harnesses 03:40 — The interoperablity hurdle 07:52 — Introducing Omnigent 10:19 — Composition, collaboration, and control 12:06 — Matei demos Omnigent 22:03 — Why open source Omnigent? Presenters: Matei Zaharia, Co-founder and CTO, Databricks