Juhee Kim

Postdoctoral researcher, Azure Research, Microsoft, Cambridge, UK.

I am a postdoctoral researcher at Azure Research, Microsoft, in Cambridge, UK. I received my Ph.D. from CompSec Lab, Department of Electrical and Computer Engineering, Seoul National University, advised by Byoungyoung Lee.

My research background spans system security from high-level user programs to low-level operating systems and hardware, with a focus on identifying new attacks and designing practical defenses. I am currently working on securing LLM-based agent systems from a system security perspective.

Publications

• DualView: Preventing Indirect Prompt Injection in Personal AI Agents
  Juhee Kim*, Woohyuk Choi*, Taehyun Kang, Youngmin Kim, and Byoungyoung Lee (* equal contribution)
  Preprint 2026
• Agent Data Injection Attacks are Realistic Threats to AI Agents
  Woohyuk Choi*, Juhee Kim*, Taehyun Kang, Jihyeon Jeong, Luyi Xing, and Byoungyoung Lee (* equal contribution)
  Preprint 2026
• SoK: Attack and Defense Landscape of Agentic AI Systems
  Juhee Kim, Wenbo Guo, and Dawn Song
  USENIX Security Symposium (Security) 2026
• DMGuard: Safeguarding Kernels from Physical Page Use-After-Free Vulnerabilities
  Juhee Kim*, Jaeyoung Chung*, Dae R. Jeong, and Byoungyoung Lee (* equal contribution)
  USENIX Security Symposium (Security) 2026
• Prompt Flow Integrity to Prevent Privilege Escalation in LLM Agents
  Juhee Kim*, Woohyuk Choi*, and Byoungyoung Lee (* equal contribution)
  Preprint 2025
• TikTag: Breaking ARM’s Memory Tagging Extension with Speculative Execution
  Juhee Kim, Jinbum Park, Sihyeon Roh, Jaeyoung Chung, Youngjoo Lee, Taesoo Kim, and Byoungyoung Lee
  IEEE Symposium on Security and Privacy (S&P) 2025
• PeTAL: Ensuring Access Control Integrity against Data-only Attacks on Linux
  Juhee Kim, Jinbum Park, Yoochan Lee, Chengyu Song, Taesoo Kim, and Byoungyoung Lee
  ACM Conference on Computer and Communications Security (CCS) 2024
• CHANCEL: Efficient Multi-client Isolation Under Adversarial Programs
  Adil Ahmad, Juhee Kim, Jaebaek Seo, Insik Shin, Pedro Fonseca, and Byoungyoung Lee
  Network and Distributed System Security Symposium (NDSS) 2021

Talks

• Bypassing ARM’s Memory Tagging Extension with a Side-Channel Attack
  Black Hat USA 2024 · Samsung Security Tech Forum 2024 · Hardware.io NL 2024

Experience

• Postdoctoral Researcher, Azure Research, Microsoft, Cambridge, UK · Jul 2026 – present
• Postdoctoral Researcher, CompSec Lab, Seoul National University · Mar 2026 – Jun 2026
• Software Engineering Intern, Meta, London, UK · Jun 2025 – Sep 2025

Education

• Ph.D., Electrical and Computer Engineering, Seoul National University · 2020 – 2026
• B.S., Electrical and Computer Engineering, Seoul National University · 2015 – 2020

Grants & Support

• Professional Grant, USENIX Security Symposium · 2026
• Ph.D. Research Scholarship, Samsung Electronics Mobile eXperience (MX) Business · Mar 2022 – Feb 2026

Academic Service

• Program Committee, ACM Workshop on Artificial Intelligence and Security (AISec) · 2026
• Artifact Evaluation Committee, USENIX Security Symposium · 2026