interview-prep

Crisp answer: Defender for Cloud is Microsoft's CSPM (Cloud Security Posture Management) and CWPP (Cloud Workload Protection Platform). It does two things: scores your security posture via Secure Score and protects workloads with threat detection and alerting.

The two pillars:

1. CSPM — Cloud Security Posture Management

Defender for Cloud continuously assesses your Azure resources against security best practices and generates recommendations:

Secure Score: a percentage (0-100%) reflecting how many security
recommendations you have implemented.

Example recommendations:
- "MFA should be enabled on accounts with owner permissions" — High
- "Storage accounts should restrict network access" — Medium
- "System updates should be installed on your machines" — Low

Each recommendation has:
- Description of the risk
- Affected resources
- Remediation steps (often a one-click fix or policy assignment)
- Impact on Secure Score if remediated

2. CWPP — Cloud Workload Protection Platform

Threat detection for specific resource types. Requires a paid plan per resource type:

Plan Protects
Defender for Servers VMs and Arc-enabled servers — MDE integration, file integrity monitoring, JIT access
Defender for Containers AKS clusters — runtime threat detection, Kubernetes audit logs
Defender for SQL Azure SQL and SQL Server — SQL injection detection, anomalous access
Defender for Storage Blob and Files — malware scanning, anomalous access patterns
Defender for Key Vault Unusual access patterns, access from suspicious IPs
Defender for DNS Detects DNS exfiltration and C2 communication

Azure Arc integration:

Defender for Cloud extends to on-premises and other cloud servers via Azure Arc. Arc-enables the server (installs an agent), which then registers it in Azure Resource Manager. Defender for Servers can then protect it exactly like an Azure VM.

# Check Defender for Cloud coverage via Azure CLI
az security assessment list --query "[].{Name:displayName, Status:status.code}" -o table

# Check Secure Score
az security secure-score list -o table

# List unhealthy recommendations
az security assessment list \
  --query "[?status.code=='Unhealthy'].{Name:displayName,Severity:metadata.severity}" \
  --output table

What to say in the interview:

"At the RHS I automated Defender for Cloud onboarding across our Azure VM and Arc-enabled server estate, reaching 100% coverage within 30 days. I used Ansible to deploy the Azure Monitor Agent and the MDE onboarding package at scale, then tracked coverage daily in the Defender inventory. I then worked through the Secure Score recommendations systematically, starting with High severity findings and using Azure Policy to enforce remediations at scale rather than fixing resources one by one."


My notes