Han Song

Song Han is an associate professor with tenure at MIT EECS and a research director at NVIDIA, where he leads the Efficient AI team. He received his PhD from Stanford University under Bill Dally and is best known for pioneering efficient AI computing techniques including Deep Compression, the Efficient Inference Engine, and more recently LLM quantization methods such as SmoothQuant, AWQ, and StreamingLLM. He co-founded DeePhi Tech, an AI chip company in Beijing that was acquired by Xilinx and is now part of AMD, as well as OmniML, which was acquired by NVIDIA.