GitHub
Adversarial ML security lab
Hands-on lab for adversarial ML attack and defense experiments.
Selected work
Research publications, security tooling, and hands-on experiments spanning 2018 to 2025.
The common thread is practical security work: understanding attack surfaces, building defenses, and testing how intelligent systems behave under pressure.
Current builds
Hands-on projects aimed at evaluating modern AI systems and turning security research into usable practice.
Research archive
Selected research across IoT security, cyber defense modeling, NLP security, and LLM-enabled systems.
Publication
LLM-driven natural-language-to-SQL workflow for HR analytics.
Publication
BERT-enhanced runtime application security for in-app threat detection.
Publication
BERT fine-tuning pipeline for HTTP payload threat classification.
Publication
Deep learning approach for forecasting terrorism risk patterns.
Publication
UAV-based cyber counterterrorism use-case exploration.
Publication
Game-theoretic model for attacker-defender control effectiveness analysis.
Publication
Automated fuzzing workflow for memory-safety and validation bug discovery.
Project
Research program on improving IoT device and deployment resilience.
Publication
IoT botnet behavior analysis with practical detection insights.