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/agile 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.

📌 Product Security Architect (Bengaluru)
🏢 Flipkart
📍 Bengaluru