AI Jailbreak & Prompt-Injection Security Expert

micro1 is seeking an AI Jailbreak & Prompt-Injection Security Expert to join a cutting-edge initiative focused on enhancing AI safety and system robustness. In this role, you will apply your specialized knowledge to stress-test next-generation AI models, helping to shape how these systems learn, reason, and perform in real-world scenarios.

Key responsibilities

• Design and implement advanced methodologies for evaluating AI system safety, specifically focusing on ethical jailbreaks, LLM red teaming, and prompt injection.
• Create comprehensive cross-domain strategies to uncover complex adversarial bypass patterns and multi-turn vulnerabilities in AI models.
• Develop and maintain regression test suites that systematically evaluate jailbreak susceptibility and potential security weaknesses.
• Construct robust evaluation frameworks to stress-test models against adversarial threats and improve overall system resilience.
• Collaborate with technical stakeholders to translate security findings into actionable improvements for model safety and risk mitigation.

Requirements

• At least 2 years of expertise in adversarial machine learning, LLM red teaming, or AI safety evaluation.
• Proven experience in researching or uncovering vulnerabilities related to ethical jailbreaks, prompt injection, or adversarial AI attacks.
• Strong background in computer science, cybersecurity, or machine learning, supported by professional or operational experience.
• Exceptional written and verbal communication skills, with the ability to document findings clearly for diverse audiences.
• Familiarity with current LLM architectures, prompt engineering techniques, and security assessment tools.

What we offer

• The opportunity to work on high-impact, cutting-edge AI safety initiatives.
• A fully remote work environment that values domain expertise and collaborative problem-solving.
• Engagement in a project that directly influences the development and security of next-generation AI systems.