Create and maintain production monitoring, alerting, observability, runbooks, support procedures, service measures, and operational dashboards covering availability, freshness, completeness, accuracy, pipeline reliability, latency, recoverability, and cost; ensure recurring incidents and temporary fixes are converted into backlog items and permanent improvements. Own overall integration job orchestration and dependency design across batch, event-driven, API, streaming, and near-real-time workloads; require dependency-based triggers, controlled parallelism, recovery procedures, failure isolation, alerting, and end-to-end schedule optimization rather than fragile time-based scheduling.