The Docking Model
The docking model is an organizational form in which humans and AI agents dock into a governed collective for a piece of work and disconnect when it is done, instead of being hired onto a team. You don’t staff a dock; you open it. Described by Serban Mogos in 2026, as part of the work of the MOGOS Collective. Read more at serban.ai/concepts/the-docking-model.
The one-line definition
The docking model is an organization where work is done by humans and AI agents that dock in and out of a single governed collective, rather than by employees on a payroll.
What it is
A traditional company is people plus tools, sized by headcount. A docked organization has a small human core, often one person, and a distributed crew of humans and agents that connect when needed and disconnect when done. The crew is governed by a control plane that decides what each docked worker may do, and observed through a cockpit where the human watches the whole thing run. The shape is “solo, but not alone.”
How it differs from the agentic organization
The consulting frame for AI at work is the “agentic organization”: humans supervising factories of agents inside the existing org chart. The docking model changes the primitive, not the seating chart. Hiring assumes a person who joins, holds a seat, and leaves. Docking assumes a worker, human or agent, who connects for a piece of work and disconnects, leaving a record. Throughput, not headcount, becomes the unit.
Why it matters
The future-of-work conversation is stuck between two bad stories: AI replaces jobs, or AI becomes your coworker. The docking model is a third: the company itself becomes a thing you dock into. It explains how one person can run a real business, what governance is for, and why this is not outsourcing: the dock keeps what it learns; every arrival makes the harbor smarter.
A concept by Serban Mogos, from the work of the MOGOS Collective. Related: The Agentariat · Autoergy.