Job Description
Job Description:
Reference Architectures: Design and enforce target-state security architectures for cloud-native, on-premise, and SaaS-based AI/ML deployments, including LLM integration patterns, retrieval-augmented generation (RAG) pipelines, and agentic workflows.
Adversarial Defense: Establish security controls against AI-specific attack vectors, including prompt injection, jailbreaking, training data poisoning, model inversion/extraction, membership inference, and supply-chain vulnerabilities in open-source models/libraries.
Data Security & Privacy: Design zero-trust data boundaries, differential privacy techniques, and strict identity & access management (IAM) controls for training datasets, vector databases, and inferencing endpoints.
MLOps/LLMOps Guardrails: Embed automated security testing (red teaming, static/dynamic analysis for ML code, output validation, shadow AI discovery) directly into CI/CD and MLOps pipelines
People Management: Direct, mentor, and build a high-performing team of AI Security Engineers and Security Architects.
Strategic Roadmap: Define the enterprise AI Security strategy, establishing metrics (KPIs/KRIs) to track risk reduction without impeding development velocity.
Framework Mapping: Map internal security controls to established standards including NIST AI RMF, ISO/IEC 42001, OWASP Top 10 for LLMs, and MITRE ATLAS.
Cross-Functional Collaboration: Partner closely with Data Science, ML Engineering, Enterprise Architecture, Legal, and Compliance teams to operationalize Responsible AI and privacy requirements (e.g., EU AI Act, GDPR).
Executive Advisory: Serve as the subject matter expert on emerging AI risks for the CISO, CTO, and executive risk boards.
Experience: 11-16 years of experience.