Hongbo Kang is an architect at Approaching.AI and holds a Ph.D. in Computer Science from Tsinghua University. His work spans large language model inference, AI infrastructure, processing-in-memory, high-performance in-memory indexing, data placement, and task scheduling. His research has appeared in SPAA, PVLDB/VLDB, The VLDB Journal, and PPoPP, with PIM-tree receiving the VLDB 2023 Best Research Paper Runner-up Award. He was a Research Fellow at the Simons Institute for the Theory of Computing at UC Berkeley. His current focus is on cross-platform LLM deployment, inference optimization, and software–hardware co-design for emerging computing platforms.