AI Security Engineer

World Wide Technology Healthcare Solutions•Jenks, OK

14h•$140,000 - $170,000•Remote

About The Position

WWT Digital’s AI-Native Security team is seeking an AI Security Engineer to build, configure, and harden AI agent security systems tailored to individual enterprise customers. Working from structured requirements delivered by our AI Security Consultants and architectural standards set by the Principal Architect, you will implement and customize the technical security controls, evaluation frameworks, and monitoring infrastructure that form our core deliverable. You will operate at the intersection of AI engineering and security engineering, translating client-specific risk profiles and deployment environments into production-ready configurations and tooling. You will work with the Digital AI delivery team and WWT Security on joint engagements targeting enterprise engineering organizations deploying agentic AI at scale.

Requirements

4–7 years of software or ML engineering experience, with at least 2–3 years focused on AI/ML security specifically
Hands-on experience with LLM application architecture: RAG, tool use, agent orchestration frameworks (LangChain, LlamaIndex, CrewAI, or similar)
Practical experience with adversarial ML techniques: prompt injection, data poisoning, model inversion, evasion attacks
NHI security implementation: service account hygiene, secrets rotation, OAuth/OIDC for machine-to-machine authentication
Security monitoring for AI in production: behavioral baselines, output anomaly detection, audit logging
Strong coding ability — Python required; comfortable with LLM SDKs, agent frameworks, and CI/CD pipeline integration
Familiarity with AI governance frameworks: NIST AI RMF, ISO/IEC 42001, emerging EU AI Act controls

Nice To Haves

red team or penetration testing background applied specifically to AI systems
experience building reusable security tooling or accelerators packaged for delivery across multiple clients

Responsibilities

Receive and interpret requirements briefs from the AI Security Consultant; translate client-specific security requirements into technical implementation plans aligned with Principal Architect standards
Implement and customize NHI (Non-Human Identity) governance frameworks for AI agents, service accounts, and API credentials based on client architecture
Build adversarial test suites for LLM-based applications: red-teaming agents, jailbreak testing, context injection, and prompt injection scenarios
Integrate security feedback loops into client AI development workflows: model evaluation gates, output monitoring pipelines, and anomaly detection
Define and implement MLSecOps practices per client requirements: model signing, provenance tracking, fine-tune data validation
Assess AI and agent system architectures for security exposure (prompt injection paths, tool misuse, data exfiltration vectors, identity sprawl) as needed to validate consultant findings
Implement AI governance and compliance controls required by regulated clients — audit trails, model cards, data lineage documentation, access logging, and bias evaluation records — translating framework requirements surfaced by the Consultant into operational, evidenceable deliverables
Develop reusable implementation patterns, configuration templates, and accelerator tooling that can be packaged into repeatable WWT offerings
Collaborate closely with the AI Security Consultant and Principal Architect to validate that delivered systems satisfy client requirements and address identified gaps