Worked on the Xilinx/llvm-aie repository to expand floating point support by enabling conversion between bfloat16 and float16 for the aie2ps target. This feature introduced native 16-bit floating point pathways, reducing translation overhead and improving efficiency for floating point workloads on AIE architectures. The technical approach involved C++ development within the LLVM framework, focusing on compiler design and embedded systems integration. By broadening floating point operation coverage, the work laid the foundation for improved performance and accuracy in future AIE deployments. No major bugs were addressed during this period, with efforts concentrated on feature development and commit-based collaboration.