In financial environments where platform teams need to consolidate signals from dozens of AWS workloads into a unified observability backend — whether Datadog, Grafana Cloud, Honeycomb, or an internal stack — the CloudWatch → OpenTelemetry bridge pattern appears as the obvious solution. But 'obvious' and 'correct' rarely coincide in architecture....