Building intelligent systems at the intersection of Cybersecurity, Machine Learning, and Network Intelligence.
Computer Science Engineering student specializing in Cyber Security with research focus in machine learning for cybersecurity, volatile memory forensics, real-time streaming engines, explainable AI, post-quantum cryptography, and network intelligence.
My work focuses on developing interpretable, reproducible, and deployment-ready machine learning frameworks for advanced threat defense.
Dual-Paradigm Binary Malware Detection Framework using Batch Machine Learning and CapyMOA Streaming Engine.
- Dual-Paradigm: Static Batch ML (LightGBM 99.98% Acc) vs CapyMOA Real-Time Streaming (ARF+Replay 99.88% Acc)
- High-Throughput: Parallel OnlineBagging processing 24,834.7 instances/second
- Cost Reduction: modAL Active Uncertainty Sampling achieving 99.83% accuracy at 20% budget (80% analyst cost saved)
- Rigorous Audit: McNemar Statistical Test (p = 0.5127 > 0.05), Brier Score Calibration, SHAP TreeExplainer, LIME Local Surrogates
- Repository
Umbrella Cybersecurity Research Collection
- Multi-dataset Intrusion Detection (UNSW-NB15, CIC-IDS2017)
- Memory Forensics & Binary Malware Analysis
- Feature Selection & Hyperparameter Optimization
- Repository
AI-Powered Network Intelligence Platform
- Real-Time Topology Discovery & Capacity Estimation
- Telemetry Traffic Analytics & Explainable Network AI
- FastAPI + React Frontend Architecture
- Repository
Post-Quantum Cryptography Defense Engine
- Zero-Allocation C++20 Kyber-768 Implementation
- Real-Time AI Introspection Guard for Side-Channel Defense
- Repository
Digital Twin Cyber-Physical Security System
- ICS/SCADA Anomaly Detection Engine
- Automated Reconnaissance & Threat Assessment
- Repository
Understanding threats, building intelligence, and securing the future through research and engineering.
Cybersecurity • Machine Learning • Research • Open Source