I am a researcher and engineer based in Seoul, South Korea, specializing in the intersection of AI, Security, and Infrastructure. My work spans from low-level system optimization to high-level deep learning applications.

• 🔭 Current Focus: Sound Source Localization, e-Commerce, and LLM-based applications.
• 🏗️ Infrastructure: I build robust development environments and cloud-native solutions, focusing on "write once, run anywhere" philosophies.
• 🔬 Research: Published in IEEE Access, Computers & Security, and Interspeech.

For a complete list, visit my Google Scholar Profile.

• SpaceNet: A Multimodal Fusion Architecture for Sound Source Localization in Disaster Response (2025) Sensors 26(1) – Proposed a fusion architecture for robust sound localization in challenging environments.
• CHAWA: Overcoming Phase Anomalies in Sound Localization (2024) IEEE Access – Addressed phase anomalies in TDoA estimation to improve localization accuracy.
• Balance, multiple augmentation, and re-synthesis: A triad training strategy for enhanced audio deepfake detection (2024) Interspeech 2024 – Introduced a triad training strategy (balancing, augmentation, re-synthesis) to improve detection generalization.
• On the defense of spoofing countermeasures against adversarial attacks (2023) IEEE Access – Analyzed vulnerabilities of spoofing countermeasures to adversarial attacks and evaluated robust defense mechanisms.

• AdMat: A CNN-on-matrix approach to Android malware detection and classification (2021) IEEE Access – Introduced a matrix-based CNN approach for highly accurate malware classification (90+ citations).
• Android fragmentation in malware detection (2019) Computers & Security – Analyzed the impact of Android ecosystem fragmentation on detection systems (43+ citations).
• Android rooting: An arms race between evasion and detection (2017) Security and Communication Networks – Presented an arms race between rooting detection and rooting evasion (52+ citations).

My research has led to several filed patents in South Korea, focusing on security systems and automated analysis methods.

• View Full Patent List on Google Patents
• Key Areas:
  • System call whitelist generation for container security.
  • Automated source-sink classification for privacy leak detection.
  • Android malware behavior signature extraction.

Languages

Cloud Native & DevOps

Data & AI

• devenv-linux: A streamlined script to set up a modern development environment on Linux systems.
• devenv-macos: My personal dotfiles optimized for macOS, featuring a setup for productivity.
• devenv-windows: Well ... no, it's actually Linux (WSL).