Byung-Gon Chun

Byung-Gon Chun is founder and CEO of FriendliAI, an AI inference platform, and a professor of computer science and engineering at Seoul National University, South Korea, currently on leave. His research group's work on continuous batching became a standard technique for serving large language models.

Moving Between Industry Research Labs for a Decade

Chun earned a PhD in computer science at the University of California, Berkeley, an MS in computer science at Stanford University, and undergraduate and master's degrees in electronic engineering at Seoul National University, according to his personal site. He was a research scientist at Intel's lab in Berkeley, California, from 2008 to 2011, working on the CloneCloud and TaintDroid projects. He then led the systems research group at Yahoo Research Silicon Valley and was a principal scientist at Microsoft before joining Seoul National University in 2013.

Publishing ORCA at Seoul National University

His lab published the ORCA paper, which introduced continuous batching and inspired the open-source vLLM engine, according to Cerebral Valley. "That work gave us a front-row seat to just how painful the inference problem really was, even for well-resourced teams," Chun said. "When we started the company, training was getting all the attention, but I knew that inference was becoming the real bottleneck."

His awards include the EuroSys 2021 Test of Time Award and the 2020 ACM SIGOPS Hall of Fame Award.