During a two-month period, Exlin contributed targeted reliability improvements to the kubernetes/kubernetes repository, focusing on backend development with Go and Kubernetes. Exlin addressed a race condition in Job deletion by ensuring immediate cleanup of backoff records, which eliminated delays in job recreation and improved the predictability of the Job lifecycle. Additionally, Exlin enhanced the Node Shutdown Manager by introducing an exponential backoff polling mechanism to refresh logind configuration after reloads, reducing the risk of shutdown failures and manual intervention. These changes, delivered through precise bug fixes and validation tests, deepened the operational resilience and stability of Kubernetes core components.