Rudy worked on enhancing deployment safety for storage workloads by updating the Helm chart configuration in the datafold/helm-charts repository. He extended the terminationGracePeriodSeconds for the storage worker from 30 minutes to 5 hours, addressing the risk of data loss during pod restarts or rolling updates. This change, implemented using YAML and leveraging his skills in DevOps and Kubernetes, ensured that storage operations had sufficient time to complete gracefully. Rudy demonstrated end-to-end ownership by making the code change, documenting it with clear commit messages, and improving traceability. The work reflected a focused, practical approach to operational reliability.