No data in monitoring dashboards due to infrastructure differences between Dev (kubeadm) and Staging (EKS) — resolved with per-environment dashboard separation and exclude configuration

Redesigned Loki, Tempo, and Thanos to store monitoring data in S3 object storage instead of PVCs, resolving AZ-binding issues on Spot instances

Only gateway appearing in Grafana dashboard — resolved by splitting Micrometer vs OTel metric paths and fixing missing namespace labels

Record of infrastructure vulnerability remediation following a 72-hour penetration test

Circular dependency between CloudTrail and S3 bucket policy during CloudTrail provisioning — resolved by pre-constructing the ARN in locals