Product Security Architect

Location: Vasanthanagar

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, jail breaking, 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.