Senior AI Security Engineer - Agentic Systems

ServiceNow; United States; Remote; USD 180,000 - 240,000; Full time

Job summary

ServiceNow is seeking a Staff AI Security Specialist to bridge AI and security for its AI-enabled products and platform. You will reason about how agentic systems break, write code to stop exploits, and collaborate with Security Engineers from prototype to production.

Responsibilities include analyzing agent security, designing runtime controls, and conducting red-team exercises to strengthen defenses as products evolve. Strong Python and AI/ML experience are required.

Qualifications

• Experience in leveraging AI capabilities into work processes or decisions.
• 6+ years combined security/engineering/research with 2+ years hands-on AI/LLM.
• Strong knowledge of agent harness security seams and tool-calling loops.
• Knowledge of memory retrieval systems and orchestration frameworks.
• Depth in at least one security domain (appsec, offensive security, identity, or detection).
• Understanding of AI threat landscape: prompt, tool, memory poisoning, cross-session leakage.

Responsibilities

• Analyze security of AI agent systems and subversion risks.
• Design and build security controls for agentic systems.
• Build prototypes end-to-end and harden them with engineering.
• Threat model new AI architectures before release and turn findings into controls.
• Red-team our own agents and defenses against prompt injection and data exfiltration.
• Develop repeatable evaluations and benchmarks for controls.

Skills

AI security

Python

Security engineering

AI/ML

Threat modeling

Runtime security

Red-teaming

Tools

MCP

Job description

ServiceNow is seeking a Staff AI Security Specialist to bridge AI and security for its AI-enabled products and platform. You will reason about how agentic systems break, write code to stop exploits, and collaborate with Security Engineers from prototype to production.

Responsibilities include analyzing agent security, designing runtime controls, and conducting red-team exercises to strengthen defenses as products evolve. Strong Python and AI/ML experience are required.