I break AI systems and web apps - with permission - then teach people how to fix them.
┌──(sayed@redteam)-[~/whoami]
└─$ whoami
Sayed Abdou
AI Red Team Expert @ Alice
M.Sc. Cyber Security · B.Sc. Computer Science
└─$ ls -la ./skills
prompt-injection jailbreaking agentic-workflows
web-pentesting api-security security-automation
red-teaming vulnerability-research teaching
└─$ cat ./mission
Help organizations discover, understand, and fix the security flaws
in their AI systems — before attackers do.
| 🧠 AI Red Teaming | 🕸️ Web & API Pentesting | ⚙️ Security Automation |
|---|---|---|
| Adversarial testing of GenAI systems & agentic apps | OWASP Top 10, bug bounty research | AI-agent-driven testing pipelines |
| Prompt injection, jailbreaks, exfiltration channels | Burp Suite, Nmap, SQLmap, Metasploit | Burp Suite over MCP, autonomous pentest orchestration |
| Tool abuse, unsafe permissions, unintended model behavior | Responsible disclosure (Bugcrowd/HackerOne) | Custom harnesses & fuzzers (Python/Bash) |
| Project | Description |
|---|---|
| ai-red-team-lab | AI red team methodology, attack taxonomy, prompt-injection fuzzer, eval scenario builder, and AI-agent security workflows — all open-source and synthetic |
| Automated Network Vulnerability Scanner | Python scanner: port scanning, banner grabbing, service detection, Nmap integration, structured reports |
| Secure Web Application Development | Reference project: authN/authZ, input validation, SQLi/XSS/CSRF mitigations, password hashing, session management |
CASA Certified API Security Analyst · ACP API Security Certified Practitioner · eJPT eLearnSecurity Junior Penetration Tester · ICCA INE Certified Cloud Associate · CCEP Certified Cybersecurity Educator Professional
I'm always open to discussing AI security, bug bounty, and automation - or a good CTF story.
"The best defense is understanding the attack."