Worked on the ROCm/AMDMIGraphX repository to deliver a performance optimization feature focused on constant propagation. The approach centered on improving the handling of literals and arguments within the graph optimization process, aiming to boost evaluation throughput and reduce runtime overhead for end users. Leveraging C++ development skills along with expertise in parallel programming and performance optimization, the work laid the foundation for faster graph optimization passes. No major bugs were addressed during this period, as the primary focus remained on enhancing efficiency and throughput. The changes contributed to more efficient evaluation and improved runtime performance for graph-based computations.