We’re looking for 5 experienced Cybersecurity Specialists for an ongoing, task-based project focused on AI red teaming and model safety evaluation.

Project Details:

• Task Duration: 1 Task = 3 hours
• Exam: 4 hours (Only experts who pass the exam will be paid)
• Start Date: Immediate

Responsibilities:

• Evaluate AI models and systems for vulnerabilities and safety issues.
• Create reproducible test cases to assess AI agent safety and capability.
• Develop automation scripts, custom tools, and environments for testing.
• Identify potential exploits or weaknesses in AI systems.

Required Skills:

• Penetration testing and offensive security (web, API, network, infrastructure).
• Scripting and automation using Python, Bash, or PowerShell.
• Containerization & CI/CD Security (Docker).
• Familiarity with AI model vulnerabilities (prompt injection, OWASP Top 10 for LLMs).
• Ability to work with LLM tools for test creation and debugging.
• Strong problem-solving and research mindset.

Desirable Skills:

• AI/ML security and red teaming experience (LLMs, AI agents, RAG pipelines).
• Knowledge of AI red-teaming frameworks (garak, PyRIT).
• Expertise in network, application, and OS security.
• Creative vulnerability research and exploit development.