Geon-Woo Kim

gwkim.jpg

Hi there! I’m Geon-Woo Kim. I’m a Ph.D. student at UT Austin working under the guidance of Prof. Aditya Akella. I received my Bachelor’s degree from Seoul National University, where I was advised by Prof. Byung-Gon Chun. My research interests lie in systems for machine learning, with a recent focus on tackling the challenges of large-scale model training and inference. Prior to beginning my Ph.D. program, I had the pleasure of working as a software engineer at Viva Republica, a startup that operates one of South Korea’s largest fintech services.

Please see my cv for more details.

Email: gwkim [at] utexas [dot] edu

Education

  • Ph.D. Student in Computer Science, 2022-Present
  • B.S. in Computer Science & Engineering, Summa cum laude
  • B.S. in Mathematical Sciences (Double Major)

Publications

2024

  1. Read-ME: Refactorizing LLMs as Router-Decoupled Mixture of Experts with System Co-Design
    Ruisi Cai, Yeonju Ro, Geon-Woo Kim, Peihao Wang, Babak Ehteshami Bejnordi, Aditya Akella, and Zhangyang Wang
    In Advances in Neural Information Processing Systems (To Appear in NeurIPS 24)
  2. Lovelock: Towards Smart NIC-hosted Clusters
    Seo Jin Park, Ramesh Govindan, Kai Shen, David Culler, Fatma Özcan, Geon-Woo Kim, and Hank Levy
    In HotCarbon Workshop on Sustainable Computer Systems (HotCarbon 24)

2022

  1. Orca: A Distributed Serving System for Transformer-Based Generative Models
    Gyeong-In Yu, Joo Seong Jeong, Geon-Woo Kim, Soojeong Kim, and Byung-Gon Chun
    In 16th USENIX Symposium on Operating Systems Design and Implementation (OSDI 22)

2021

  1. Terra: Imperative-Symbolic Co-Execution of Imperative Deep Learning Programs
    Taebum Kim, Eunji Jeong, Geon-Woo Kim, Yunmo Koo, Sehoon Kim, Gyeongin Yu, and Byung-Gon Chun
    In Advances in Neural Information Processing Systems (NeurIPS 21)

2019

  1. Apache Nemo: A Framework for Building Distributed Dataflow Optimization Policies
    Youngseok Yang, Jeongyoon Eo, Geon-Woo Kim, Joo Yeon Kim, Sanha Lee, Jangho Seo, Won Wook Song, and Byung-Gon Chun
    In 2019 USENIX Annual Technical Conference (USENIX ATC 19)

2017

  1. Pado: A Data Processing Engine for Harnessing Transient Resources in Datacenters
    Youngseok Yang, Geon-Woo Kim, Won Wook Song, Yunseong Lee, Andrew Chung, Zhengping Qian, Brian Cho, and Byung-Gon Chun
    In Proceedings of the Twelfth European Conference on Computer Systems (EuroSys 17)