Phishing Detection with Machine Learning

Built a phishing URL classifier using feature engineering (URL length, entropy, domain reputation) and ML models (Random Forest, Logistic Regression), achieving 95% detection accuracy on test data.

Network Intrusion Detection System

Developed a deep learning–based IDS (CNN + LSTM) to classify anomalous traffic patterns; improved zero-day attack detection rates compared to traditional signature-based systems.

Active Directory Attack Simulation Lab

Designed a red-team home lab using Kali Linux + Windows Server AD to simulate Kerberoasting, lateral movement, and credential harvesting attacks; integrated Wazuh + ELK Stack for centralized logging, monitoring, and alerting.