Joo-Young Kim is the Founder and CEO of HyperAccel, a South Korean fabless semiconductor startup pioneering the world's first LLM Processing Unit (LPU) designed exclusively for large language model inference.

Kim earned his Ph.D. from KAIST (Korea Advanced Institute of Science and Technology) and spent nine years at Microsoft working on chip development at Microsoft Research and Azure, before returning to KAIST as an Associate Professor of Electrical Engineering.

In 2022, Kim presented research at the Hot Chips conference describing a specialized processor for running language models, predating the broader industry adoption of the LPU concept by over a year.

Kim leads HyperAccel's partnership with SEMIFIVE and Samsung Foundry to mass-produce the Bertha AI inference chip on a 4nm process, with Naver Cloud and global Big Tech companies as target customers.

HyperAccel has raised over $180M across Series A and Series B rounds, with the Series B led by Krafton pushing the company's valuation to 800 billion KRW, as Bertha enters full-scale mass production.