Course / Level 5
Level 5
Operable
Agents can run internal workflows — not just customer-facing actions.
Level 5 is the inside view: at least one core workflow — intake, scheduling, quoting, triage, processing — is substantially operated by agents, with human oversight, documented handoff rules, and metrics that distinguish agent work from human work. It comes with real governance: unique agent identities with human sponsors, end-to-end observability, tamper-evident logs, model governance policies, and enforced oversight boundaries.
This is the leading edge of the standard — and clearly where operations are heading. It's an ongoing program rather than a one-time project, so the businesses that start now build a real head start; the curriculum is drawn from live engagements as the patterns settle.
This level’s lessons are part of the course.
These lessons need an active enrollment. One payment unlocks every level — and Level 0 stays free for everyone. Already enrolled? Log in to pick up where you left off.
5 steps in this level
-
Step 1
A production AI workflow
Standing up at least one real operational workflow that agents run, with human oversight.
-
Step 2
Agent identities & sponsors
Unique agent identities, each with a human sponsor, in the corporate directory.
-
Step 3
Observability & audit
OpenTelemetry GenAI instrumentation, end-to-end tracing, and tamper-evident audit logs.
-
Step 4
Governance & oversight
Model governance, SLAs with error budgets and circuit breakers, and enforced human oversight boundaries.
-
Step 5
Recap
What changed, why it matters, and how to package this level into a service you can sell.