Worked on the ROCm/rocm-systems repository to enhance GPU performance monitoring by exposing energy count and current socket power metrics to profiling tools such as PAPI. Leveraged C++ and system programming skills to update device dependency maps, ensuring that GPU metrics dependencies are accurately reflected for performance analysis tools. This feature enables more precise energy-aware profiling and supports optimization efforts across ROCm tooling. The work focused on integrating new metrics into existing performance monitoring frameworks, improving the visibility of GPU power consumption for developers and researchers. No bugs were addressed during this period, with efforts concentrated on feature development and integration.