GenAI Cyber Security Expert

This is a fully remote position, open to applicants in Illinois, +1 more state.

📋 Description

• Conduct adversarial testing to enhance the safety guardrails of GenAI against significant cyber threats.

• Perform thorough red-teaming exercises on GenAI models within SCADA, ICS, and CBRNE contexts.

• Investigate potential safeguard breaches, dangerous capability escalations, and inadvertent exposure of sensitive OT data.

• Create adversarial prompts, jailbreaks, and multi-turn assessments tailored for industrial hacking situations.

• Assess model manipulation involving payload creation, ladder logic exploitation, reconnaissance, or the aggregation of harmless inputs into risky workflows.

• Record failure modes, categorize their severity, and communicate findings to safety and alignment teams.

• Develop repeatable testing suites, adversarial datasets, and risk taxonomies.

• Keep abreast of GenAI attack methodologies and CBRNE/critical infrastructure threats.

• Assist in Alice's AI safety and security services, encompassing model hardening assessments, pre-deployment red-teaming, runtime guardrails, and drift detection.

⛳️ Requirements

• Must reside in the US and possess work authorization, excluding Illinois and Texas.

• An advanced degree in Cybersecurity, Computer Science, Engineering, Security Studies, CBRNE sciences, or a related discipline, or equivalent professional, military, or government experience.

• Extensive expertise in industrial cybersecurity, particularly with SCADA, ICS/OT, industrial protocols, PLCs, and cyber-physical attack chains.

• Solid knowledge of CBRNE, especially regarding its integration with cyber and digital systems.

• Experience in red-teaming for penetration testing, adversarial simulations, vulnerability research, or offensive security.

• Experience or aptitude in GenAI red-teaming, including adversarial prompts, jailbreak research, multi-turn testing, and LLM safety evaluations.

• Practical exposure to SCADA/OT hacking, including vulnerability research, protocol fuzzing, and network architecture analysis.

• An adversarial mindset for detecting safeguard circumvention through decomposition, obfuscation, or multi-turn manipulation.

• Strong technical writing skills for documenting methodologies, findings, and mitigation strategies.

• Preferred: hands-on experience in LLM red-teaming, AI safety evaluations, or jailbreak/prompt research.

• Preferred: experience in creating red-team test cases, datasets, benchmarks, or evaluation pipelines.

• Preferred: familiarity with frontier AI safety principles, dangerous capability assessments, and responsible scaling frameworks.

• Preferred: background in offensive security, threat intelligence, incident response, or adversary emulation.

• Preferred: knowledge of the MITRE ATT&CK framework (including ICS), NIST frameworks, and IEC 62443.

• Preferred: experience in national security, defense, intelligence, or high-consequence industrial settings.

• Preferred: relevant offensive-security or ICS/OT certifications.

• Preferred: extensive expertise in GenAI cyber red-teaming, SCADA/ICS/OT red-teaming, chemical/industrial cyber risk, biological/lab cyber risk, radiological/nuclear cyber risk, or cyber-physical/high-consequence threat analysis.

🏝️ Benefits

• Competitive salary and performance-based bonuses.

• Comprehensive health, dental, and vision insurance.

• Opportunities for professional development and continuing education.

• Flexible work hours and remote working options.

• Collaborative and innovative work environment.