AI Security Architect – Generative AI, LLM Security & AI Gateway Controls

Axiom Path•Charlotte, NC

21d•Remote

About The Position

The AI Security Architect will help advance cybersecurity capabilities that support the secure adoption of artificial intelligence within a highly regulated financial services environment. The role is part of a global AI security transformation initiative establishing consistent safeguards for generative AI applications, intelligent automation, and emerging AI technologies. It involves hands-on work translating AI security risks into technical controls, scalable architecture patterns, and sustainable security practices, in collaboration with international cybersecurity, engineering, risk, compliance, technology, and business teams. The position designs and implements security architectures and centralized AI gateway controls for generative AI applications, LLM integrations, copilots, and agent-based systems. It addresses risks including prompt injection, sensitive data exposure, insecure model outputs, unauthorized AI behavior, vulnerabilities in RAG and vector database architectures, model dependencies, and AI supply chains. The architect also supports security integration across development and operations, monitoring and audit capabilities, shadow AI assessment, and the creation of reusable security platforms and operational processes. This is a 100% remote W2 contract opportunity supporting a global AI security transformation program. The role collaborates with teams across the Americas, Europe, and Asia-Pacific.

Requirements

5–7 years of relevant cybersecurity, security architecture, or security engineering experience, including hands-on involvement in securing modern applications and AI-enabled technologies.
Practical experience designing or implementing security controls for generative AI, LLM-powered applications, AI copilots, or agentic AI systems.
Strong understanding of AI-specific threats, including prompt injection, data leakage, insecure output handling, excessive agent permissions, and model supply chain vulnerabilities.
Experience designing or implementing centralized security controls for AI model interactions, API traffic, or application gateways.
Knowledge of prompt filtering, response inspection, PII and secret redaction, access controls, rate limiting, and security audit logging.
Familiarity with security considerations involving RAG architectures, embeddings, vector databases, and external model integrations.
Proficiency in at least one programming or scripting language, such as Python, Go, or Node.js.
Experience integrating security capabilities into software development, application architecture, and production environments.
Understanding of data protection and security requirements in environments handling sensitive or regulated information.
Ability to work independently across architecture, engineering, implementation, and operational adoption.
Strong communication and collaboration skills, including the ability to explain technical risks and architectural decisions to diverse stakeholders.

Responsibilities

Design, develop, and implement security architectures for enterprise generative AI applications, LLM integrations, copilots, and agent-based systems.
Establish technical safeguards against prompt injection, sensitive data exposure, insecure model outputs, and unauthorized AI system behavior.
Architect centralized AI gateway capabilities that enforce security policies across interactions with AI models and services.
Implement prompt inspection, output filtering, sensitive information redaction, model authorization, and API abuse prevention controls.
Address AI security vulnerabilities involving RAG, vector databases, embeddings, model dependencies, and AI supply chains.
Partner with software engineering and platform teams to integrate security controls throughout AI application development, deployment, and ongoing operations.
Develop secure architectural patterns for AI systems processing confidential customer information, employee records, and regulated financial data.
Support identification and assessment of unauthorized or unmanaged AI tools, or shadow AI.
Establish monitoring, audit logging, and evidence-generation capabilities for AI security oversight and regulatory readiness.
Collaborate with regional security leaders and risk management, compliance, privacy, legal, procurement, and technology stakeholders.
Translate security assessments into actionable implementation plans, measurable controls, and repeatable operational processes.
Build automation, integrations, and security tooling using modern programming or scripting languages.