Unit of value
Cost per token is an input metric. Cost per accepted investigation or incident is the decision metric.
Attribute model and tool consumption to each incident and identify actionable cost drivers.
Cost per token is an input metric. Cost per accepted investigation or incident is the decision metric.
Join provider usage to trace, incident, team, environment and outcome.
Include retries, failed runs, observability and supporting infrastructure.
Apply per-run limits, daily anomaly thresholds and environment budgets.
Reduce unnecessary work before switching models: context, loops, caching, routing and tool precision.
For every arrow, the engineer must name the data, identity, trust boundary, failure mode and evidence produced.
AI and documentation are allowed. The learner owns every generated artifact and must preserve the stated safety boundaries.
Symptom → impact → hypothesis → evidence → test → root cause → containment → correction → verification.
Propose budget controls that reduce runaway cost without suppressing high-value investigations.
Record the achieved L-level only from observed evidence. Course completion and certification do not automatically change the skill matrix.