Juntaek Lim

Korea Advanced Institute of Science & Technology (KAIST) [Feb. 2025 ~ Present]

Ph.D. in School of Electrical Engineering

Advisor: Prof. Minsoo Rhu

Korea Advanced Institute of Science & Technology (KAIST) [Sep. 2022 ~ Aug. 2024]

M.S. in School of Electrical Engineering

Advisor: Prof. Minsoo Rhu

Korea Advanced Institute of Science & Technology (KAIST) [Mar. 2018 ~ Aug. 2022]

B.S. in School of Electrical Engineering

Minor in School of Computing

Summa Cum Laude

Undergraduate Research Intern at Vertically Integrated Architecture Research Group (VIALab), KAIST

Dec. 2021 ~ Aug. 2022

• • DNN accelerator: implementation of cycle-level simulator of SCNN & analysis of various DNN accelerators

  • Mixture of Experts (MoE): analysis of prior work related to MoE including GShard code

KAIST EE Co-op Internship at Samsung Electronics, Hwaseong

Mar. 2021 ~ Aug. 2021

• Study of AXI protocol

• Study of the tiling scheme and dataflow used in Exynos NPU

• RTL design and synthesis for fetching unit (FU) of Exynos NPU

Juntaek Lim, Gungyu Lee, Zirui Neil Zhao, Mattan Erez, and Minsoo Rhu, "DynamicCC: Efficient Confidential MoE Inference on Trusted CPU-GPU Serving Systems," The 33rd International Symposium on High-Performance Computer Architecture (HPCA-33), Salt Lake City, UT, Mar. 2027

• [Paper] [Code]

Yunjae Lee*, Juntaek Lim*, Jehyeon Bang, Eunyeong Cho, Huijong Jeong, Taesu Kim, Hyungjun Kim, Joonhyung Lee, Jinseop Lim, Ranggi Hwang, Se Jung Kwon, Dongsoo Lee, and Minsoo Rhu, "Debunking the CUDA Myth Towards GPU-centric AI Systems," The 52nd IEEE/ACM International Symposium on Computer Architecture (ISCA-52), Tokyo, Japan, Jun. 2025

• [Paper] [Code]

Taehun Kim, Yunjae Lee, Juntaek Lim, and Minsoo Rhu, "A Characterization of Generative Recommendation Models: Study of Hierarchical Sequential Transduction Unit," IEEE Computer Architecture Letters (CAL), Feb. 2025 

• [Paper]

Juntaek Lim, Youngeun Kwon, Ranggi Hwang, Kiwan Maeng, Edward Suh, and Minsoo Rhu, "LazyDP: Co-Designing Algorithm-Software for Scalable Training of Differentially Private Recommendation Models," The 29th ACM International Conference on Architectural Support for  Programming Languages and Operating Systems (ASPLOS-29), San Diego, CA, USA, Mar. 2024

• • Acceptance Rate: 20% (81 among 409)

  • [Paper] [Code]

• Korea Presidential Scholarship in Science and Technology (대통령 과학 장학금) (2025)

• MICRO-58 Artifact Evaluation Committee

• Teaching Assistant, EE312: Introduction to Computer Architecture, Spring 2024 @KAIST

• Teaching Assistant, EE209: Programming Structure for Electrical Engineering, Fall 2023 @KAIST

• Teaching Assistant, EE312: Introduction to Computer Architecture, Spring 2023 @KAIST

• Programming Languages: Python, C, C++, CUDA, Verilog HDL, TPC-C

• Machine Learning Frameworks: PyTorch