Principal Security Engineer – AI & Copilot Data Protection

Tasks and responsibilities

• This role serves as a leading technical authority on AI privacy, promotes critical thinking, generates actionable insights, provides guidance on the use of the Microsoft AI Copilot platform, and shares expertise with peer teams.
• Area of Responsibility (Principal Practitioner):
  • Tier 2 technical consulting and monitoring strategy for AI and Copilot data protection, not just for enterprise functions.
  • Definition of robust, repeatable patterns to validate AI model workflows that can be adopted by others.
  • Close collaboration with product managers, compliance, and AI innovation teams (product, information security, data protection, audit, and GRC teams).
  • Advising C-suite executives and senior leadership teams to ensure that risks associated with Microsoft AI are appropriately mitigated.
  • Serve as a management consultant for templates and controls in AI security assessments to ensure corporate compliance.
• AI and Copilot Security Architecture:
  • Serves as a trusted technical leader and design authority for AI model and enterprise AI security controls (Large Language Models (LLM), MLOps, and RAG).
  • Definition of secure-by-design architectural models for deploying AI models to ensure the protection of AI-related models and vector databases.
  • Research and evaluation of commercial third-party models, including:
    • Risk Assessment of Vendor Benchmarks
    • Supply Chain Integrity Analysis
    • AI-Related Security and Privacy Metrics
    • Transparency, verifiable findings, and human oversight
  • Providing security blueprints for the development of AI models, behavior, data pipelines, and full-stack deployment.
• Control Engineering and Operations:
  • Design, implementation, and operation of automated controls, consisting of:
    • Strategy for Information Disclosure and Alerting
    • Self-configuration of AI-based content protection and AI acceptance for use
    • Shadow Data Management and Data Protection Compliance
    • Data Journey Management and Pipeline Observability
  • Uniform, structured observability.
  • Development of monitoring, alerting, and dashboards to support the adoption of high-risk AI usage patterns.
  • Develop proactive measures that enable the detection of automated violations and improve governance.
  • Documentation of standards and tools, regular reviews, and ongoing consultations.
• Governance and Institutionalization:
  • Providing AI insights from internal abuse analyses and benchmarking security controls.
  • Collaborating with governance teams to provide support:
    • Adoption of AI Innovations
    • Documentation of Quality and Usage
    • Desktop Usage and Error Handling
    • Safe House and Adoption Reports
    • Candidate and AI Model Deployment
  • Maintenance of Sector Patterns in Copilot Monitoring, Demo-Trust