Some more about me (҂◡_◡) ᕤ

I’m Rayane Boucheraine, also known as R4Y4N3, an Offensive AI Security Researcher and Red Teamer based in Algeria.

My work focuses on identifying and exploiting security vulnerabilities in LLM-powered applications, autonomous agents, and multi-system AI workflows. I’m particularly interested in prompt injection (PI/IPI), MCP and tool-use security, RAG vulnerabilities, authorization boundaries, and sensitive data exposure.

I conduct independent security research and participate in AI bug bounty programs, including 0DIN, and adversarial red-teaming challenges on Gray Swan Arena. My approach involves analyzing attack surfaces, developing reproducible proofs of concept, and understanding how vulnerabilities can lead to unintended agent behavior or unauthorized actions.

I primarily work with Python, Rust, Go, C, and JavaScript, using Linux and various security research tools.

I’m always interested in exploring new attack surfaces, challenging the security assumptions of agentic systems, and collaborating on meaningful AI security research.

For AI security research, bug bounty hunting, red teaming, or technical collaborations, feel free to reach out.

Email me · Download CV · Read the blog

Contact

• GitHub: github.com/R4Y4N3
• LinkedIn: rayane-boucheraine
• Email: r_boucheraine@estin.dz