AI Agents Just Became Multiplayer. Here’s How to Build Them.

AI Agents Just Became Multiplayer. Here’s How to Build Them.

More

Summary

Most people use AI as a single-player tool, with each teammate building their own skills, integrations, and agent setup. This video from Nate Herk looks at a different model: a shared workspace where humans and AI agents collaborate in the same place, using a tool called Kylon.

The interface resembles Slack, with channels on the left and chats on the right, but each channel can hold apps, automations, files, and knowledge. Team members can talk to the same agent at once, and agents can message each other. Users can also bring their own Claude Code and Codex subscriptions into the workspace, so several people and several agents work on one project, such as building and restyling a dashboard.

Herk walks through onboarding a new workspace, including inviting teammates, creating agents, and giving an agent context about your business by exporting a markdown brief from an existing AI operating system. He explains the pricing model, which charges per workspace rather than per member, and how credit usage is tracked on the team plan. He discloses that Kylon approached him to cover the tool and addresses his initial objections about price. A useful look at where team-based AI workflows may be headed.


๐Ÿ“บ Source: Nate Herk | AI Automation ยท Published October 15, 2026
๐Ÿท๏ธ Format: Tutorial Demo

1 Item

Channels