Charan Ponnada expanded the performance testing capabilities of the pytorch/pytorch repository by integrating AMD AWS runners into the continuous integration pipeline. Using Shell and YAML, Charan developed architecture-specific configurations that enabled the CI system to run targeted performance benchmarks on multiple AMD platforms, addressing the need for broader hardware coverage. The work focused on CI/CD automation and cloud computing, with updates to both infrastructure documentation and configuration files to support the new runners. By improving the feedback loop for performance tests on AMD hardware, Charan accelerated the identification of regressions, delivering a focused and technically deep feature within the month.