Building AI-powered systems that detect and stop cyber threats before they cause damage. Offensive security + machine learning.
I'm Ahmed Mahmoud, a Penetration Tester and AI-Driven Cybersecurity Engineer focused on offensive security, vulnerability research, and intelligent threat detection systems.
My work combines artificial intelligence with cybersecurity — designing ML-powered security tools that analyse network behaviour, detect anomalies, and assist analysts in identifying threats in real time.
Currently pursuing my B.Eng. in Network & Cybersecurity at Elsewedy University while actively hunting vulnerabilities on HackerOne and building research-grade security tooling.
Actively hunting vulnerabilities across public and private bug bounty programs on HackerOne. Focusing on web application vulnerabilities including IDOR, XSS, SSRF, authentication bypass, and business logic flaws.
Working on information security assessments, vulnerability research, and security tooling within a cybersecurity-focused environment.
Developing and training ML models for cybersecurity applications. Working with classification, anomaly detection, and sequence models applied to network traffic and threat analysis.
Participated in a Microsoft-backed summer camp focused on cloud and security technologies, delivered through Sprints' intensive program structure.
Intensive hands-on cybersecurity training covering vulnerability scanning, web application security, cloud security, and SIEM operations at Egypt's National Telecom Institute.
AI-powered NDR for Industrial Control Systems that classifies OT network traffic in real time using a TensorFlow autoencoder for anomaly detection and a secondary classifier distinguishing reconnaissance from actual attacks — enabling operators to triage threats without manual packet inspection.
Hybrid AI botnet detection combining ensemble learning (LightGBM + XGBoost + Isolation Forest + DBSCAN) with SHAP explainability and real-time BiLSTM + GRU + Attention sequential analysis.
Real-time IDS combining network forensics and ML that detects malicious traffic using Suricata + XGBoost trained on CIC-IDS datasets, with Zeek enrichment and a live Streamlit alert dashboard.
End-to-end GRC framework for M&A cyber risk: due diligence questionnaire → quantitative risk scoring → integration decision framework → post-merger governance tracking.
Linux iptables-based firewall with dynamic IP/port/protocol blocking, persistent SQLite logging, real-time stats, and an interactive CLI — replacing manual iptables with a structured rule lifecycle.
Looking to collaborate on cybersecurity research, red team engagements, or AI-driven threat detection? I'm open to internships, freelance work, and research partnerships.
I typically respond within 24 hours.