Jiin Kim
Jiin Kim
Embedded Files
I am a Ph.D. student in the Vertically Integrated Architecture Research Group at KAIST, advised by Prof. Minsoo Rhu.
My research interests span computer architecture, systems, and large-scale AI infrastructure. I focus on improving the efficiency and scalability of large language model (LLM) serving and agent-based AI systems, with particular emphasis on inference scheduling, memory optimization, and hardware-software co-design. More broadly, I aim to design future AI systems that balance efficiency, accuracy, and scalability. If you are interested in my research, please feel free to contact me :)
Email: jiin.kim@kaist.ac.kr
Office: N1 818 @ KAIST
Education
Education
Ph.D in School of Electrical Engineering, KAIST
Mar. 2024 ~ Current
Advisor: Prof. Minsoo Rhu
M.S in School of Electrical Engineering, KAIST
Mar. 2022 ~ Feb. 2024
Advisor: Prof. Minsoo Rhu
B.S. in School of Electrical Engineering, KAIST
Mar. 2017 ~ Feb. 2022
Double major in Computer Science
Cum Laude
Experience
Experience
Undergraduate research intern at VIA Research Group (Prof. Minsoo Rhu), KAIST
Jun. 2020 ~ Dec. 2020 / Jun. 2021 ~ Feb. 2022
Accelerator for CNN
Distributed computing / Microservice
Undergraduate research intern at LANADA Lab (Prof. Yung Yi) , KAIST
Dec. 2019 ~ Feb. 2020
Reinforcement Learning
Intern at SK Hynix, Bundang
Jun. 2019 ~ Aug. 2019
Department of SoC, Analog Circuit
Publication
Publication
Jiin Kim, Byeongjun Shin, Jinha Chung, and Minsoo Rhu, "The Cost of Dynamic Reasoning: Demystifying AI Agents and Test-Time Scaling from an AI Infrastructure Perspective"
[arXiv]
Under review
Jiin Kim*, Gwangoo Yeo*, Yujeong Choi, and Minsoo Rhu, "PREBA: A Hardware/Software Co-Design for Multi-Instance GPU based AI Inference Servers"
[arXiv]
Under review
Yujeong Choi, Jiin Kim, and Minsoo Rhu, "ElasticRec: A Microservice-based Model Serving Architecture Enabling Elastic Resource Scaling for Recommendation Models," The 51st International Symposium on Computer Architecture (ISCA-51), June 2024
[Paper]
* Co-first authors with equal contributions
Skills
Skills
Python, C++, C, Kubernetes
Page updated
Google Sites
Report abuse