Open to opportunities

Rohit Prasanth

@rohitprasanth

Message

Cybersecurity researcher and CVE author specializing in offensive AI/LLM security and web application exploitation.

India

Message

What I'm looking for

I’m looking to apply adversarial research and GenAI security skills to break and secure real-world AI systems—pairing offensive testing with responsible disclosure and remediation collaboration.

I’m a cybersecurity researcher, CVE author, and offensive security practitioner focused on finding and fixing real-world weaknesses across AI/LLM systems and web applications. I’ve authored high-impact disclosures like “CVE-2026-45553” (CVSS 8.7 High) and work closely with upstream maintainers to ensure vulnerabilities get patched.

My work centers on adversarial red-teaming—using techniques like prompt injection, jailbreak methods, and AI model behavior analysis to evaluate GenAI attack surfaces. I’ve responsibly disclosed critical security issues affecting national infrastructure and coordinated remediation with the relevant security teams.

I also bring deep hands-on offensive capability through CTF work and exploit-driven research. I’ve solved 100+ CTF challenges and authored offensive security challenges, while mentoring others in vulnerability research and responsible disclosure practices.

Alongside research, I build security tooling—like a Mobile Vulnerability Scanner for Android APK intent exploitation and a modular Web Application Security Scanner for OWASP Top 10 detection. I’m currently seeking to apply adversarial research and GenAI security skills to break and secure real-world AI systems.

Experience

Work history, roles, and key accomplishments

Current

Independent Security Researcher

Current

Independent

Jan 2023 - Present (3 years 5 months)

Conducted AI/LLM adversarial red-teaming and responsible disclosure (e.g., CVE-2026-45553, CVSS 8.7) across Qwen/DeepSeek models and enterprise web apps, coordinating remediation with affected security teams. Reported findings to DataCamp, Grok, and Notion and assessed prompt-injection and adversarial-input attack surfaces.

AI Red Teaming Prompt Injection Web Application Security Python Threat Modeling

Current

Security Researcher & CTF Player

Current

Team Bi0s

Jan 2023 - Present (3 years 5 months)

Solved 100+ CTF challenges across web exploitation, cryptography, and binary exploitation, and authored offensive security challenges for national and international competitions. Mentored junior members in vulnerability research, offensive techniques, and responsible disclosure practices.