Worked on the NVIDIA/aistore repository to deliver ETL Pod Template Standardization and Startup Simplification, focusing on improving deployment consistency for ETL workloads in Kubernetes environments. Developed Python-based server execution paths and adjusted container port configurations to ensure reliable and reproducible ETL job startups. Embedded the new pod templates directly into the codebase, streamlining onboarding for new teams and supporting future enhancements. The work emphasized DevOps practices and ETL orchestration, leveraging Python and YAML to manage Kubernetes pod templates. This approach reduced startup variability, tightened CI/CD processes, and enabled more predictable ETL runtimes across different deployment environments without introducing new bugs.