Bruce Chen developed and integrated DeepEP Round-Robin Expert Routing for low-latency inference in the IBM/vllm repository. His work focused on enabling efficient expert placement by implementing a round-robin scheduling strategy, which improved both latency and resource utilization for DeepEP model deployments. Bruce introduced new methods for managing expert routing tables and modified existing functions to support the updated routing logic. Using Python, PyTorch, and his expertise in deep learning and model optimization, he laid the groundwork for scalable expert routing across deployments. The work was feature-driven, with no major bugs addressed, reflecting a focused and technically deep engineering contribution.