Developed and integrated the SiMM storage backend for HiCache within the sgl-project/sglang repository, enabling distributed key-value caching with RDMA acceleration. This work focused on backend development and leveraged Python to implement efficient caching mechanisms across distributed systems. By introducing RDMA support, the integration improved both latency and throughput for HiCache, laying the foundation for scalable, production-ready deployments. The approach emphasized seamless backend integration and code provenance, ensuring maintainability and traceability. No major bugs were addressed during this period, with efforts concentrated on delivering a robust new feature that enhances cache scalability and performance in distributed environments.