Every agent system decides how much can happen without a person, but many teams never make that decision explicitly. They default to one of two extremes: approve every step or let the agent run freely. Neither is right for every task.

This talk introduces five autonomy levels—operator, collaborator, consultant, approver, and observer—and treats autonomy as a per-task design choice rather than a platform-wide setting. In a live refund-approval demo, the same request runs at three levels. The comparison exposes what changes in speed, permissions, evidence, recovery, and the cost of a mistake.

Attendees will leave able to explain the five levels and their tradeoffs, match autonomy to a task’s risk profile, design permission and approval boundaries, and identify where an existing agent is over- or under-autonomous. The goal is not maximum autonomy; it is deliberate, observable control.

Audience: Engineers and engineering leaders designing agents that can take actions.
Format: 60–75-minute framework talk with a live comparative demo.
Demo: One refund request executed at three autonomy levels with different permission, evidence, and approval boundaries.
Evidence: New session; currently in evaluation for Visual Studio Live! Las Vegas 2027.
Materials: Talk-specific assessment worksheet, repository, slides, and recording are not yet published.
Vendor scope: Vendor-neutral design framework.