The shift toward the cloud-native paradigm and distributed, microservices-based systems has established Kubernetes as the de facto standard for container orchestration. However, as organizations increasingly operate fleets exceeding 20 clusters to support mission-critical workloads, the complexity of lifecycle management—specifically infrastructure upgrades—has scaled exponentially. Within this landscape, Liqo provides a peer-to-peer model for transparent workload offloading and resource sharing. Despite its utility, current Liqo upgrade procedures remain entirely manual, requiring administrators to follow exact, error-prone sequences that risk leaving systems in inconsistent, non-functional states. This thesis designs, implements, and validates a Kubernetes operator that automates Liqo upgrades with minimal service disruption. The proposed solution employs a three-stage infrastructure upgrade architecture—sequencing Custom Resource Definition (CRD) migration before control plane updates, and control plane updates before network fabric modifications—to respect inherent architectural dependencies.