AI Security Architect

Role: AI Security Architect

Engagement type: Contract

Seniority: Senior individual contributor

Location: Remote (Eastern or Central Time Zones), local to Charlotte, NC would be preferred

Anticipated duration: until December 31, 2026, with potential extension and/or FTE conversion

Clearance / eligibility: Must be authorized to work in the United States

Role Summary

We are seeking a senior Security Architect specializing in AI security to serve as the technical authority for the secure adoption of artificial intelligence across enterprise and product environments. The contractor will define how AI/ML pipelines and model supply chains are secured, govern the safe use of generative AI and LLM-integrated applications, secure AI-assisted development practices, and design defenses against AI-specific threats such as prompt injection, data poisoning, model theft, and unsafe agentic workflows.

While AI security is the core of this engagement, the contractor is expected to apply broad security architecture expertise to the cloud, identity, and data platforms on which AI capabilities depend. Success requires deep technical expertise, strong communication skills, and the ability to influence technical and business stakeholders without direct authority.

Key Responsibilities

• Own AI security architecture strategy and standards, mapping controls to recognized AI risk frameworks (e.g., NIST AI RMF, OWASP LLM Top 10, MITRE ATLAS, Google SAIF) while ensuring alignment with broader enterprise security standards (NIST CSF, CIS 18, Zero Trust principles), business strategy, and risk appetite.
• Develop threat models for AI/ML systems, establish guardrails for generative AI usage (including LLM-integrated applications and AI-assisted development), and evaluate AI security tooling.
• Design security architecture for AI agents and agentic workflows, including least-privilege access to data and systems, human-in-the-loop controls, output validation, and monitoring for prompt injection and model abuse.
• Partner with legal and privacy functions to operationalize responsible and compliant AI, translating emerging AI regulation and standards (e.g., EU AI Act, ISO/IEC 42001, NIST AI RMF) into architectural requirements and AI risk assessments.
• Design and govern AI-related security controls for cloud environments (Azure, AWS), including identity architecture, network segmentation, data protection, and workload security posture.
• Provide authoritative security architecture reviews for new products, platforms, infrastructure, and third-party integrations, delivering actionable findings and remediation guidance.
• Develop and maintain security reference architectures, patterns, and guardrails for use by engineering and development teams; socialize and evangelize standards across technology organizations.
• Partner with security analysts and engineers to ensure architectures are informed by threat actor TTPs and emerging attack vectors.
• Translate architectural decisions into control evidence, supporting audits, client assessments, and regulatory requirements.
• Evaluate and recommend security tooling and vendors; contribute to build-vs-buy decisions and support procurement processes as a technical subject matter expert.
• Mentor security engineers and cross-functional technology peers.

Required Qualifications & Experience

• 8+ years of progressive cybersecurity experience, with at least 3 years in a security architecture or senior engineering capacity, including recent demonstrated experience securing AI/ML or generative AI systems in production.
• Experience securing enterprise AI platforms and copilots (e.g., Azure OpenAI, Microsoft Copilot, GitHub Copilot, Amazon Bedrock, Google Vertex AI) and integrating AI telemetry into SIEM/SOC monitoring.
• Demonstrated expertise in enterprise security architecture, including Zero Trust design, network segmentation, identity and access management, and cloud-native security patterns.
• Hands-on experience securing AI/ML systems: threat modeling for AI pipelines, data poisoning and prompt injection mitigations, and governance frameworks for generative AI adoption (e.g., OWASP LLM Top 10, MITRE ATLAS).
• Familiarity with AI-specific security controls and tooling such as LLM guardrails, prompt/response filtering, AI gateways, model red-teaming, and AI security posture management (AI-SPM).
• Strong working knowledge of major public cloud platforms (Azure, AWS), including cloud security posture management (CSPM), SIEM, SSE/SASE, and endpoint security.
• Experience developing security architecture documentation and reference architectures.
• Proven ability to influence cross-functional technology teams and communicate complex security concepts to both technical and non-technical stakeholders.
• Strong analytical, critical thinking, and written communication skills; self-directed and able to manage competing priorities in a fast-moving environment.

Desired Qualifications

• Security certifications such as CISSP, CCSP, AWS/Azure Security Specialty, or GIAC credentials.
• Experience building or red-teaming AI agents and LLM-integrated applications.
• AI security / governance credentials such as ISACA AAISM/AAIA, IAPP AIGP, or ISO/IEC 42001 Lead Implementer (or equivalent demonstrated expertise).
• Exposure to product security and secure SDLC.
• Experience with lightweight automation or scripting (Python, PowerShell, or similar) to support architecture validation and tooling integration.
• Familiarity with relevant frameworks and standards. Examples: NIST CSF, NIST AI RMF, CIS 18, ISO 27001, SOC 2, and applicable privacy regulations.
• Working knowledge of the AI/ML technology stack and its security implications: model lifecycle (training, fine-tuning, inference), RAG and vector databases, MLOps/LLMOps pipelines, model and prompt registries, and API-based model consumption.
• A bachelor s degree in a related technical field or equivalent professional experience.

Prioritizing Guidance

To prioritize candidates, please screen for the following. Strong candidates will clearly evidence the first three.

• Concrete, hands-on delivery of AI/GenAI security controls in a production environment (not solely policy or advisory work).
• Ability to articulate specific AI threats (e.g., prompt injection, data poisoning, model theft, insecure agentic tool use) and the controls that mitigate them.
• Security architecture depth in at least one major cloud, with identity and data protection experience.
• Comfort operating as an individual contributor and influencing without direct authority.