Donghwan Kim

Donghwan Kim

PhD student @ Penn State

About Me

Hi, I am a 3rd year PhD student in Computer Science and Engineering at The Pennsylvania State University, advised by Prof. Kiwan Maeng. My research interest lies in building efficient computer systems through hardware and software optimization. My work has spanned privacy-preserving machine learning (differentially private training, fully homomorphic encryption) and extends to LLM serving systems. Before joining Penn State, I was fortunate to work at SCALE lab, advised by Prof. Jung Ho Ahn.

(Updated. Sep 2026)

News

Our paper “Characterizing How Complex Agentic AI Systems Handle General Tasks: A Trace-Based Simulation Study” has been accepted to IISWC’26 and selected as a Best Paper Finalist!

Our recent work “Cocoon: A System Architecture for Differentially Private Training with Correlated Noises” was accepted to OSDI’26!

Publications

  • [IISWC’26] Donghwan Kim, Prakhar Singh, Younghoon Min, Jongryool Kim, Jongse Park, Kiwan Maeng, “Characterizing How Complex Agentic AI Systems Handle General Tasks: A Trace-Based Simulation Study”, in IEEE International Symposium on Workload Characterization (IISWC), 2026. (Best Paper Finalist)

  • [OSDI’26] Donghwan Kim, Xin Gu, Jinho Baek, Timothy Lo, Younghoon Min, Kwangsik Shin, Jongryool Kim, Jongse Park, Kiwan Maeng, “Cocoon: A System Architecture for Differentially Private Training with Correlated Noises”, in USENIX Symposium on Operating Systems Design and Implementation (OSDI), 2026. [link]

  • [CAL’25] SeokHyeon Kong, Donghwan Kim, Euiseong Seo, Kiwan Maeng, “Characterizing the System Overhead of Discrete Noise Generation for Differential Privacy”, in IEEE Computer Architecture Letters (CAL), 2025. [link]

  • [CCS’24] Jae Hyung Ju*, Jaiyoung Park*, Jongmin Kim, Donghwan Kim, Jung Ho Ahn, “NeuJeans: Private Neural Network Inference with Joint Optimization of Convolution and FHE Bootstrapping”, in ACM SIGSAC Conference on Computer and Communications Security (CCS), 2024. (*equal contribution) [link]

  • [Access’23] Donghwan Kim*, Jaiyoung Park*, Jongmin Kim, Sangpyo Kim, Jung Ho Ahn, “HyPHEN: A Hybrid Packing Method and Its Optimizations for Homomorphic Encryption-based Neural Networks”, in IEEE Access, 2023. (*equal contribution) [link]

  • [DISCC’23] Jaiyoung Park, Donghwan Kim, Wonkyung Jung, Sangpyo Kim, Jongmin Kim, Jung Hee Cheon, Jung Ho Ahn, “Toward Practical Privacy-Preserving Convolutional Neural Networks Exploiting Fully Homomorphic Encryption”, in Workshop on Data Integrity and Secure Cloud Computing (DISCC), 2023. [link]

  • [ISCA’23] Jongmin Kim, Sangpyo Kim, Jaewan Choi, Jaiyoung Park, Donghwan Kim, Jung Ho Ahn, “SHARP: A Short-Word Hierarchical Accelerator for Robust and Practical Fully Homomorphic Encryption”, in International Symposium on Computer Architecture (ISCA), 2023. [link]

Education

The Pennsylvania State University, State College, PA — Aug 2024 – Present

  • Doctor of Philosophy, Computer Science and Engineering
  • Advisor: Kiwan Maeng

Seoul National University (SNU), Seoul, Korea — Mar 2022 – Feb 2024

  • Master of Science, Interdisciplinary Program in Artificial Intelligence
  • Advisor: Jung Ho Ahn

Seoul National University (SNU), Seoul, Korea — Mar 2016 – Feb 2022

  • Bachelor of Science in Electrical and Computer Engineering (*2-years leave due to military service)

Work Experience

LLM Serving System Research Intern – SK Hynix America, San Jose, CA — May 2026 – Aug 2026

Tutorials

  • “CrypTorch: Easy-to-use and Efficient PyTorch-based Compiler Infrastructure for Machine Learning with MPC”, ASPLOS 2026 Tutorial [link]

Teaching Experience

Introduction to Computer Architecture (CMPEN 431) — Fall 2025

  • Weekly recitation, grading assignments

Computer Organization and Design (CMPEN 331) — Fall 2024

  • Weekly recitation and office hours; grading assignments and exams

Programming Methodology — Spring 2022

  • Taught weekly lab sessions / Implemented C++ final term project