Overview

• Design and implement an intelligent framework for automated document classification, labelling, and protection using AI and Data Loss Prevention (DLP) policies.
• The project aims to enhance data governance, ensure compliance with regulatory standards (e.g., GDPR), and prevent unauthorized data exposure across Microsoft 365 and other platforms.
• Number of interns: 01
• Project Ref: CED-BI/SECURITY-002

Objectives

• Automate and scale document classification and sensitivity labelling based on content analysis.
• Enforce real-time protection rules to prevent data leaks and unauthorized sharing.
• Improve compliance posture and audit readiness by consistent application of labels and DLP policies.

Key Components & Technologies

• AI models for content-based classification and sensitivity detection (AI/ML frameworks).
• Data cataloging and labelling integrated with Microsoft Information Protection and Azure Information Protection.
• DLP Policies enforced across Microsoft 365 using Microsoft 365 Compliance Center and Microsoft Graph API.
• Implementation and automation using PowerShell and Python scripts.

Tasks & Deliverables

• Research and select or train AI/ML models for document/content sensitivity detection.
• Integrate classification models with a data catalog and automated labelling workflows.
• Define and implement DLP policies and rules in Microsoft 365 Compliance Center; automate policy deployment via Microsoft Graph API.
• Provide scripts, documentation, and a demonstrable prototype showing automated classification, labelling, and enforcement.

Required Skills & Knowledge

• Hands-on experience with Microsoft Information Protection, Azure Information Protection, and Microsoft 365 Compliance Center.
• Familiarity with AI/ML frameworks for NLP/content classification and with data governance tools.
• Scripting and automation experience in PowerShell and Python; experience using Microsoft Graph API.
• Understanding of data protection regulations (e.g., GDPR) and DLP concepts.

Expected Outcomes & Evaluation

• Automated and scalable document classification pipeline with consistent sensitivity labelling.
• DLP policies and enforcement mechanisms that provide real-time protection against data leaks.
• Improved audit readiness and demonstrable metrics (e.g., classification accuracy, number of prevented exposures, policy coverage).
• Deliverables include code repository, deployment/automation scripts, design documentation, test results, and a final project report/presentation.