Worked on the llm-d/llm-d repository to deliver an inference scheduling performance enhancement by enabling dshm memory configuration. Focused on optimizing resource utilization and reducing latency under load, the work involved updating the values_cpu.yaml configuration file using YAML to leverage dshm memory for improved inference scheduling efficiency. Applied skills in Kubernetes, configuration management, and infrastructure management to implement and validate these changes. Local testing and configuration verification ensured that the update did not introduce regressions in inference workloads. The approach demonstrated a methodical focus on performance optimization through targeted configuration adjustments within a Kubernetes-managed infrastructure environment.