DORA, SLOs, and incidents

Running it after you ship is the back half of the platform engineer's job. Key points:

• DORA dashboards are built from your deployment and incident data. The four metrics are surfaced from the delivery and incident events you feed in, not computed from a survey and not invented by the platform.
• SLOs: set the objective per service, and the platform tracks the error budget against it and warns when burn climbs, so reliability is a decision, not a surprise. Each objective lives on the service's catalog entry.
• Incidents keep detection, the change that caused it, and the recovery in one record, and link back to the deploy, so the change failure rate reflects what really happened.