Security Architect, Agentic AI
Core
Designing and securing agentic AI systems (LLM apps that plan, use tools, and manage memory) to protect enterprise data and prevent cyberattacks.
Role type
Senior Security Architect / Security Engineer (Agentic AI)
Builds
Secure agent runtimes, libraries, MCP servers, tool registries, and reference implementations for IT and Product teams.
Domain
Cybersecurity / Artificial Intelligence / Cloud Security
Deliverable
production ML models
Required skills
Agent runtime design (planning, tool calling, memory controls), Python/TypeScript/Node, Cloud security, Kubernetes security, LLM API integration, RAG security, Threat modeling, Prompt injection defense, Data exfiltration prevention
Preferred skills
Advanced guardrail architectures, Real-time alignment enforcement, OWASP LLM Top 10, MITRE ATLAS, NIST AI RMF, Red-team testing harnesses, Retrieval authorization
Technologies
LangGraph, LangChain, Semantic Kernel, Kubernetes, MCP (Model Context Protocol)
Responsibilities
Build secure agent runtimes and reference implementations; Secure MCP servers, clients, and tool registries with auth/audit; Enforce secure-by-default controls (schema validation, allowlists, redaction); Threat model and test agent workflows for prompt injection and data exfiltration.
Seniority
Senior, hands-on IC