I will perform ai and llm red teaming and jailbreak vulnerability testing

S
sk_automation33
S
sk_automation33
Saeed Ahmed
Sommige informatie wordt in het Engels weergegeven.
perform ai and llm red teaming and jailbreak vulnerability testing
Volledig scherm
perform ai and llm red teaming and jailbreak vulnerability testing

Over deze dienst

Is your AI chatbot or LLM app safe from jailbreaks and prompt injection? I test it manually, the way a real attacker would, before a real attacker does.


What I test:

- Jailbreak and safety-bypass attempts

- Prompt injection (direct and indirect)

- System prompt and data leakage

- Multi-turn manipulation and misuse cases


What you get:

- A written PDF report with each finding, the exact prompt used, a risk rating (Low/Medium/High) and steps to reproduce

- Fix recommendations (Standard and Premium packages)


I have documented and responsibly reported LLM vulnerability findings to a major AI vendor. All testing is manual, confidential, and limited to systems you own or have permission to test.


Please message me before ordering so we can confirm the scope.


  • Modelexpertise

    • Generatieve AI

  • Branche

    • Cyberbeveiliging

  • Programmeertaal

    • Python

  • Taal

    • Engels

  • Technische expertise

    • AI-ethiek en beperking van vooroordelen

Maak kennis met Saeed Ahmed

Saeed Ahmed

AI Red Teaming Specialist for LLM Security and Vulnerability Testing

  • Afkomstig uitPakistan
  • Lid sindsmei 2026

  • Talen

    Urdu, Engels
I test AI systems (chatbots, LLM-integrated apps, AI agents) for safety and security vulnerabilities before attackers or bad actors find them. I specialize in adversarial prompt testing, jailbreak/prompt injection testing, and document fraud resistance testing. Recent work: independent research testing 4 major AI models across a structured 4-tier adversarial framework, with findings responsibly disclosed to each provider.

Mijn portfolio

Gerelateerde tags