AI Security Associate Manager

The AI Security Associate will work within the Cyber Risk team in the Information Security organization. 

The Work

• Validate compliance with internal AI security, Responsible AI, and data protection requirements using established frameworks and checklists and output from AI testing tools. 
• Document findings clearly, including non-compliance findings and recommended mitigations.
• Monitor updates to AI platforms, models, and frameworks to ensure test coverage remains current.
• Support continuous improvement by contributing to process and review enhancements, reusable scripts, and updated security review criteria.

What you need

• Strong understanding of GenAI ecosystems, including LLMs, vector databases, prompt frameworks, agent orchestration, and multi agent workflows.
• Working knowledge of AI-specific attack vectors such as prompt injection, data leakage, model inversion, jailbreaking, and unsafe agentic behavior.
• Familiarity with AI security testing tools, red team prompt testing suites, agent behavior analyzers, and LLM evaluation frameworks.
• Ability to evaluate AI system behavior for hallucinations, unauthorized data access, harmful output, and misuse scenarios.
• Understanding of cloud security fundamentals across Azure/AWS/GCP to interpret AI system context (identity, logging, boundary protections). 
• Ability to review and evaluate AI prompt configurations, system inputs, guardrails, and testing outputs for risks and misconfigurations.
• Basic scripting skills (Python, JS, or similar) to create reusable test cases, automation scripts, and AI test harnesses.
• Familiarity with Responsible AI principles, including transparency, fairness, safety, and explainability.
• Ability to apply AI security and RAI requirements using frameworks, checklists, and system documentation, similar to those referenced in enterprise AI standards.
• Ability to analyze agentic or autonomous system behavior to detect policy violations, unsafe actions, or decision making risks.
• Ability to identify non-compliance findings, unsafe outputs, ambiguous prompts, or risky automations—and clearly articulate why they matter.
• Skill in assessing the security impact of new AI features, updated model versions, or platform changes.
• Ability to perform light threat modeling of AI-enabled workflows, including attack chains involving agents, tools, and integrations.
• Strong technical writing ability to document test results, findings, and recommended mitigations in a clear and actionable format.
• Strong communication skills to share AI security testing findings to engineers, product teams, and leadership.
• Skill in explaining complex AI system risks in plain language for non technical audiences.
•  Ability to work independently and as part of a high-performing team.
• Ability to continuously improve skills and staying current with changes in AI models, platforms, and  emerging threats—tracking updates and integrating them into test coverage.
• Experience contributing to updates of AI security patterns, guidance, and internal testing frameworks"