Summary
Junxue Zhang is a professor and researcher with 13 years of experience specializing in high-performance, intelligent, and secure datacenter networking, now focusing on networked systems for LLMs. He earned a PhD from HKUST’s iSINGLab and transitioned from Research Assistant Professor at HKUST—where he contributed to SIGCOMM and OSDI papers on RDMA and GPU interconnects—to academia at USTC. Previously as CTO of Clustar he built full-stack privacy-preserving computation platforms, including FPGA and ASIC accelerators for federated learning and homomorphic encryption, with multiple systems papers at NSDI, EuroSys and ATC. His work bridges cutting-edge systems research and practical hardware/software co-design, driving innovations in RDMA reliability, multi-NIC GPU communication, and privacy-preserving ML acceleration. Based in Hefei, he is actively recruiting PhD/MSc students, postdocs, and RAs to advance LLM networking and secure datacenter infrastructures.
13 years of coding experience
5 years of employment as a software developer
Master's degree, Computer Science, Master's degree, Computer Science at Southeast University
Hong Kong University of Science and Technology (HKUST)