Functional Manager Cyber Security
ICICI Securities · Banking & Financial Services
- NAVI MUMBAI, India
- On-site
- Posted today
- Apply by 30 Oct
- Cybersecurity
About the job
AI Cyber Security Architect
Key Responsibilities
Lead client engagements focused on securing AI, GenAI, and ML workloads across AWS, Microsoft Azure, and Google Cloud.
Design and implement AI security strategies, architectures, governance models, and control frameworks aligned with business and regulatory requirements.
Conduct AI security assessments across cloud-hosted AI/ML and GenAI environments.
Evaluate controls against NIST CSF, CSA CCM, ISO 27001, NIST AI RMF, ISO 42001, and related frameworks.
Lead threat modelling and architecture reviews for AI and GenAI solutions.
Assess risks including prompt injection, jailbreaks, insecure output handling, excessive agent permissions, sensitive data exposure, model misuse, and insecure tool integrations.
Define secure reference architectures for model endpoints, APIs, RAG pipelines, vector stores, plugins, agent frameworks, orchestration layers, and external integrations.
Guide secure deployment of Amazon Bedrock, SageMaker, Azure AI Foundry, Azure OpenAI, Azure Machine Learning, and Google Vertex AI.
Implement controls around private connectivity, encryption, key management, secrets management, access restriction, tenant isolation, content safety, logging, and monitoring.
Integrate security into MLOps, and MLSecOps pipelines.
Embed policy validation, secrets handling, model governance, and compliance checks into CI/CD and ML workflows.
Define controls across the full AI lifecycle, including data ingestion, training, validation, deployment, runtime monitoring, incident response, drift review, and retraining.
Lead AI governance activities covering model inventory, risk classification, approval workflows, traceability, control mapping, audit readiness, and policy alignment.
Drive AI security posture management using CNAPP, CSPM, CWPP, DSPM, SIEM, and cloud-native security tools.
Oversee monitoring and incident response for suspicious prompt activity, agent misuse, unauthorized model access, data leakage, and control failures.
Lead AI security testing including misuse-case testing, prompt attack validation, model guardrail reviews, red-team support, and remediation planning.
Manage stakeholders across Security, Cloud, DevOps, Architecture, Data, ML Engineering, Privacy, Legal, and Risk teams.
Contribute to proposals, solutioning, thought leadership, reusable assets, and cloud AI security practice development.