InfoSec Penetration Tester for AI Training

Apply hands-on cybersecurity expertise to help train next-generation AI systems through high-quality, real-world security input. This remote contract role supports an enterprise SaaS project, with no prior AI experience required.

• Design comprehensive red-team workflows and adversarial scenarios for SaaS and cloud environments.
• Conduct penetration testing and adversarial threat modeling across Microsoft 365, Google Workspace, Slack, GitHub, and Snowflake.
• Serve as a senior reviewer, assessing the quality and depth of red-team exercises and deliverables.
• Identify and document AI-agent misuse and manipulation risks, including prompt injection, privilege escalation, data exfiltration, and destructive actions.
• Provide practical input on security controls, policies, and detection strategies for real-world SaaS environments.
• Collaborate with cybersecurity professionals and provide detailed written and verbal feedback on findings and improvements.
• Create adversarial scenarios for benchmarking that reflect current threats and attack methods.

• Extensive hands-on experience in red teaming, penetration testing, application security, cloud security, or security engineering.
• Deep understanding of enterprise SaaS security, including identity, RBAC and permissions, and data governance.
• Security evaluation experience with Microsoft 365, Google Workspace, Slack, GitHub, and Snowflake.
• Ability to assess adversarial ways AI and SaaS platforms could be exploited or compromised.
• Strong written and verbal communication skills for explaining complex security issues and recommended remediations.
• A demonstrated record of practical, real-world impact on security projects.
• Experience in AI or LLM security, or adversarial testing of agent-based systems, is a bonus.

• Remote contract engagement.

$60 to $100 per hour.