Worked on the ytsaurus/ytsaurus-k8s-operator repository to enhance the Kubernetes Operator’s upgrade process for YTSaurus clusters. Focused on developing a robust rolling update mechanism for Execution Nodes, the work consolidated the update flow by removing deprecated fields and defaulting the update plan to a no-op when empty. Using Go and Kubernetes, the developer expanded test coverage to validate update behavior and prevent regressions, while also streamlining code for maintainability and backward compatibility. These improvements reduced upgrade risks in production environments and demonstrated a strong emphasis on deployment lifecycle reliability, maintainability, and the practical application of API development and DevOps skills.