Worked on enhancing benchmarking workflows and deployment processes across the llm-d/llm-d and llm-d/llm-d-benchmark repositories. Delivered a new configuration and results guide for the gpt-oss-120b model, updating workload parameters and deployment instructions to streamline multi-model benchmarking and improve onboarding. Addressed cross-platform compatibility by updating file copy commands to support both kubectl and OpenShift CLI, removing unsupported flags to reduce deployment friction. Leveraged skills in Kubernetes, OpenShift, and Python, and utilized YAML and Markdown for documentation and configuration. The work focused on improving benchmarking accuracy, deployment efficiency, and repeatability, contributing to more robust machine learning infrastructure.