null null
JOB SUMMARY
The successful applicant will be responsible for evaluating and mitigating technical risks in AI and Machine Learning (ML) models and Web based AI platforms. This role anchors the organization’s AI Governance agenda ensuring AI is deployed in a secure, compliant, ethical, and auditable manner aligned to enterprise risk appetite.
Assessing the security posture of existing and proposed AI systems, platforms, and processes to protect and continually improve the confidentiality, integrity, and availability of the AMLI’s business.
KEY RESPONSIBILITIES
AI Model Governance Risk Assessment
Validate core technical design, architecture, and code of production-ready AI/ML models.
Assess models for algorithmic bias, data drift, concept drift, and data lineage gaps.
Maintain a centralized inventory of all enterprise AI assets and risk profiles.
AI Security Vulnerability Management
Lead AI-focused VAPT / Red Teaming:
Prompt injection, jailbreak testing
Model extraction and evasion attacks
Data poisoning and adversarial testing
Extend traditional AppSec practices (SAST/DAST) to:
LLM pipelines
RAG systems
Agent workflows
- Build AI testing playbooks aligned to OWASP AI Testing standards for trustworthiness validation
MCP (Model Context Protocol) Agentic AI Security
- Assess and mitigate risks inMCP-based architecture:
- Credential aggregation / single point of failure risks
- Tool poisoning, schema injection, and command execution risks
- Confused deputy and excessive privilege scenarios
- Define security standards for:
- AI-to-tool integrations
- Identity, authentication, and authorization models
- Ensure adherence toleast privilege and secure integration practices
AI Threat Intelligence Control Enhancement
- Track emerging threats across:
- AI supply chain and model dependencies
- Agentic AI risks and automation misuse
- Translate intelligence into:
- Detection rules (SOC/XDR)
- Preventive guardrails (prompt filtering, output validation)
MEASURES OF SUCCESS
Compliance with internal AI security standards and regulatory requirements
Reduction in critical AI vulnerabilities (e.g. Prompt Injection)
KEY RELATIONSHIPS (INTERNAL /EXTERNAL)
Business Functions, Third Parties and Control groups
KEY COMPETENCIES/SKILLS REQUIRED
VAPT, Configuration Reviews, Network Architecture reviews, Ethical Hacking
Experience
Duration: 3 to 5 years in Machine Learning, model validation, or AI risk.
Domain: Prior experience in BFSI, Fintech, or Insurance is highly preferred.
Certifications (Preferred): OSCP, CISSP, CAIPT (AI certification)
KEY TECHNICAL SKILLS
Languages: Python, R, SQL.
Frameworks: TensorFlow, PyTorch, MLflow.
XAI Tools: SHAP or LIME
Cloud: AWS, Azure, or Google Cloud AI governance tools.
