Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
Agentic Kanban: Making AI workflows behave like real teams
Learn how Agentic Kanban structures AI agents like a team, moving tasks across a board for research, execution, and QA, enabling collaborative refinement and iteration.
Agentic Kanban is a system that uses a Kanban-style interface to coordinate multiple AI agents as a structured workflow rather than a single loop.
Instead of one agent doing everything, tasks move across a board and are picked up by role-specific agents (e.g. research, execution, QA), enabling iteration and review.
In the demo, I’ll show how a high-level goal is broken into tasks, how agents collaborate through shared state, and how outputs are refined across multiple steps.
The focus is on making agent systems behave more like real teams, including where this approach works and where it still breaks.