Fangcheng has broad interests in distributed machine/deep learning algorithms and systems and geo-distributed federated learning. He has published more than 20 papers in top academic conferences and journals such as SIGMOD, VLDB, SOSP, ICML, NeurIPS, KDD, and TKDE. He is the designer and/or program committee (PMC) member for several system projects, including Hetu, PowerFL, and Angel. His works in these system projects have been featured in several billion- or even trillion-scale machine/deep learning applications in industries.