Design and execute adversarial testing campaigns against AI models, including large language models and autonomous agents.
This role is remote.
Responsibilities
Research attack vectors and prompt injection techniques to identify vulnerabilities, jailbreaks, and unintended behaviors.
Conduct red team exercises simulating real-world deployment scenarios to probe for bias, toxicity, and misinformation.
Create adversarial datasets and benchmarks to evaluate model robustness under various attack conditions.
Collaborate with security teams to document findings and provide remediation recommendations.
Required Skills
5+ years of experience in penetration testing, red team operations, or AI Red Team activities.
Proficiency with testing tools such as SPLX, Garak, TextAttack, PyRIT, Burp Suite, Metasploit, Nessus, Cobalt Strike, NMAP, or sqlmap.
Understanding of AI, Machine Learning, cloud computing, software applications, and networking.
Degree in computer science, information technology, programming, development, artificial intelligence, or cybersecurity from an accredited institution or equivalent work experience.
Preferred Skills
Background in offensive cyber operations or related fields.