AI Jailbreak & Prompt‑Injection Security Specialist (Remote)

About the role

We are seeking a highly skilled AI security expert to join a leading technology client on a contract basis. The role focuses on uncovering and mitigating prompt‑injection and jailbreak vulnerabilities in advanced AI models, ensuring robust protection across the deployment pipeline.

Key responsibilities

• Conduct deep‑dive analyses of AI model weaknesses related to prompt‑injection and jailbreak techniques.
• Design, implement, and execute custom jailbreak scenarios to test model resilience.
• Develop and refine testing methodologies for evaluating AI security mechanisms.
• Document findings, produce detailed reports, and recommend concrete security improvements.
• Collaborate with engineering, product, and DevOps teams to embed security best practices into AI deployment workflows.
• Stay up‑to‑date with emerging AI threat vectors and evolving attack methods.

Required profile

• Proven experience in AI security, specifically prompt‑injection and jailbreak testing.
• Strong background in adversarial machine learning and ethical hacking.
• Ability to analyze security logs and interpret model behavior under attack.
• Experience with threat modeling and risk assessment for AI systems.

Required skills

• Python programming and scripting for automated security testing.
• Familiarity with AI model architectures and inference pipelines.
• Knowledge of prompt‑injection testing frameworks and jailbreak tools.
• Expertise in adversarial machine learning techniques.
• Competence in security log analysis, threat modeling, and risk assessment.

What we offer

• Remote work from anywhere.
• Competitive contract rate of $50‑$90 per hour.
• Opportunity to influence security strategy for a global technology leader.