Cyber Security Specialist

Overview

AI-savvy freelancers

prompt-based or web-based attacks

What You’ll Do

• Create and run

  adversarial tests

  on AI systems (e.g., prompt injections, indirect attacks, or contextual exploits).
• Build

  test scenarios

  that show how an AI can be tricked into unintended actions or responses.
• Document how you did it — steps, reasoning, and outcome — in clear, reproducible reports.
• Suggest improvements or mitigations for discovered weaknesses.
• Stay updated on

  AI safety and red teaming trends

  .

Skills Needed

• Strong understanding of

  AI prompts

  ,

  LLM behavior

  , and

  autonomous agents

  .
• Experience in

  AI red teaming

  ,

  ethical hacking

  , or

  adversarial testing

  .
• Familiarity with

  web technologies

  (HTML, JavaScript) and tools like

  Docker

  or

  Python

  .
• Creative problem-solving and ability to “think like an attacker.”
• Clear, structured documentation and reporting skills.
• READ TEAM

• Experience : 6+ years

Bonus Skills

• Experience testing or breaking systems like

  AutoGPT

  ,

  Computer Use

  , or

  Claude / ChatGPT Agents

  .
• Knowledge of

  cybersecurity fundamentals

  (XSS, phishing, sandbox bypass, etc.).
• Prior work in

  AI safety or adversarial ML research

  .

Why This Gig

• Work on

  cutting-edge AI security challenges

  from anywhere.

• Flexible, project-based pay

  — rewarded per validated and documented attack.
• Great opportunity to

  showcase your AI security and creative hacking skills

  .

Email: sajid.ahmed@truelancer.c

omContact: +91-9064877846 (WhatsApp onl