Worked on the zalando-incubator/kubernetes-on-aws repository to address a critical observability issue affecting new Kubernetes clusters on AWS. Delivered a targeted fix by updating the default metrics endpoint and port configuration using YAML, ensuring that metrics collection remained stable during cluster upgrades and rollouts. Applied configuration management and DevOps practices to resolve configuration drift, which previously caused metric collection failures and impacted observability reliability. Leveraged Git-based change management to implement and document the solution, reducing incident risk and supporting service level agreements. This work improved the reliability of observability tooling and accelerated the deployment of new Kubernetes clusters in production environments.