AI Agent Security Architect
Core
Define long-term security architecture and technical patterns for autonomous agents, MCP integrations, multi-turn reasoning, and multi-agent coordination.
Role type
Senior IC AI Agent Security Architect
Builds
Runtime guardrails, semantic firewalls, intent verification filters, policy-enforcement systems, and secure micro-isolated execution environments for agent code and tool use.
Domain
AI/ML Security, Agentic Frameworks, Cloud Security
Deliverable
production ML models | infrastructure
Required skills
Agentic architecture design, AI threat modeling, OWASP Top 10 for LLMs, NIST AI RMF, MITRE ATLAS, Python/Go/TypeScript, code review, security tooling development, runtime sandboxing design, technical documentation.
Preferred skills
Firecracker, gVisor, Docker, WebAssembly, MCP, LangChain, AutoGen, CrewAI, ReAct frameworks.
Responsibilities
Architect and deploy runtime guardrails and policy-enforcement systems; conduct agentic threat modeling and lead AI red-team initiatives; design secure execution environments for agent code and tool use; secure vector databases and RAG pipelines; maintain authoritative AI security design patterns and SDKs; establish tamper-evident logging and telemetry standards; translate AI security controls into enterprise audit documentation.
Seniority
Senior, hands-on IC