AI Security Engineer
Core
Design adversarial attack scenarios, secure agentic systems, and develop secure-by-default reference architectures to protect next-generation enterprise AI deployments.
Role type
Senior individual contributor AI security engineer
Builds
Secure agentic systems, automated security guardrails, and secure AI reference architectures
Domain
Enterprise AI security, Agentic AI, Cloud Security
Deliverable
production ML models
Required skills
AI red-teaming, adversarial attack scenario design, Python, MLOps tooling, AI frameworks (PyTorch, TensorFlow), agent protocols (MCP), cloud security architecture (AWS, Azure, GCP), OWASP Top 10 for LLMs, NIST AI RMF standards
Preferred skills
Commercial AI security platforms, shadow AI discovery tools, MITRE ATLAS tactics, ISO/IEC 42001, EU AI Act, open-source AI security contributions
Technologies
PyRIT, Garak, Promptfoo, MCP, AWS, Azure, GCP
Responsibilities
Assess and secure AI agents, tool integrations, and agentic workflows; Design and execute advanced adversarial attack scenarios including prompt injection and jailbreaking; Partner with AI engineers to establish automated security guardrails and output filtering policies; Audit AI Ops pipelines and vet third-party AI supply chain components; Develop enterprise AI security standards and train cross-functional teams
Seniority
Senior, hands-on IC