Worked on the Mooncake repository to implement cache-aware interfaces within MasterService, enabling the scheduler to access detailed cache metadata for improved resource allocation and request routing. Designed and delivered new APIs that allow retrieval of all cache keys, segments, and precise queries on segment capacity and usage, reducing scheduling guesswork and supporting scalable scheduling strategies. Leveraged Go for interface design and integrated advanced caching concepts to enhance system observability. Focused on distributed systems and system programming principles, this work established a foundation for future enhancements in cache metrics collection, supporting more data-driven and efficient scheduling decisions across the platform.