Developed the YuanrongConnector for distributed inference within the vllm-project/vllm-omni repository, focusing on enabling scalable multi-node communication using the Yuanrong Datasystem. The connector was architected atop the OmniConnectorBase framework, emphasizing modularity and reusability for future extensions. This work laid the technical foundation for higher throughput and more efficient distributed deployments, aligning with best practices in backend development and distributed systems design. Leveraging Python and data engineering skills, the implementation addressed the need for robust connector-based architectures in large-scale inference scenarios. No major bug fixes were recorded during this period, with efforts concentrated on feature delivery and system extensibility.