Learning Guides · June 2026 · 13 min read


Azure is the cloud platform that dominates enterprise. Over 95% of Fortune 500 companies use it. It has the deepest Microsoft 365 integration on the market. And in the UK and Europe, it is the default choice for public sector and regulated industries.

If you want a cloud engineering career in 2026, Azure is one of the most commercially valuable places to focus. This roadmap takes you from zero knowledge to job-ready in five months — with a project-based approach that builds a portfolio, not just a certificate collection.


Why Azure in 2026

Before committing five months to anything, you deserve to understand the market.

SignalFigureSource
Azure market share (enterprise)24% global, dominant UK/EU enterpriseSynergy Research 2025
Azure-skilled engineer demandConsistently top 5 most-requested cloud skillsRobert Half 2026
UK DevOps/cloud engineer shortage70%+ of employers struggling to hireLorien UK 2025
Azure DevOps + AKS roles (UK lead salary)£80,000–£110,000Lorien UK Tech Salary Survey 2025
Employer willingness to pay above-market for Azure expertise87% of tech leaders facing skills shortageRobert Half 2026

The gap between supply and demand is wide. Azure-skilled engineers who can also work with Kubernetes, Terraform, and CI/CD are being hired aggressively.


How This Roadmap Works

Five months. One project per month. Each project compounds on the last until you have a portfolio that proves you can build, operate, and explain production-grade Azure infrastructure.

Certifications are optional side effects — not the goal. The goal is employability, and that comes from demonstrable skill, not badges.


Month 1: Foundations

Cloud fundamentals

Start with the conceptual layer before any Azure-specific tooling. Understanding these concepts means you can reason about any cloud, not just Azure.

  • Compute: virtual machines, containers, serverless
  • Networking: IP addressing, subnets, routing, load balancing, DNS
  • Storage: object storage, block storage, databases, caching
  • Identity: authentication vs authorisation, roles, policies
  • The shared responsibility model

Azure fundamentals

ConceptAzure Service
Virtual machinesAzure Virtual Machines
Containers (managed)Azure Container Apps, AKS
Serverless functionsAzure Functions
Object storageAzure Blob Storage
Relational databaseAzure SQL, Azure Database for PostgreSQL
Identity and accessMicrosoft Entra ID (Azure AD), RBAC
NetworkingAzure Virtual Network, NSGs, Azure DNS
MonitoringAzure Monitor, Log Analytics

Month 1 project: Three-tier web application

Deploy a web app, app tier, and managed database on Azure using the portal — then immediately re-do it using the Azure CLI. Document the architecture with a diagram.

This forces you to understand what the portal is doing under the hood. You will reference this project for the rest of the roadmap.


Month 2: Infrastructure as Code

Manual deployments do not scale and do not pass code review. Everything gets automated from Month 2 onwards.

Azure CLI and PowerShell

Learn to provision resources without touching the portal. These are the building blocks before moving to full IaC.

  • Resource groups, tags, subscriptions
  • RBAC assignments at resource and subscription scope
  • Policy assignments and compliance checks

Terraform on Azure

Terraform is the industry-standard IaC tool. Learning it on Azure is transferable to AWS and GCP.

Core concepts:

  • Azure provider configuration and authentication
  • Resources, data sources, variables, outputs
  • Remote state with Azure Blob Storage backend
  • Modules for reusable infrastructure patterns
  • terraform plan, apply, destroy — and why plan output matters before any apply

Azure-native IaC: Bicep

Bicep is Microsoft's native IaC language. It compiles to ARM templates and has first-class Azure tooling support. Worth knowing — especially for organisations already invested in the Azure ecosystem.

Month 2 project: Full infrastructure automation

Re-deploy your Month 1 three-tier app entirely with Terraform. No portal. No manual steps. Store state in Azure Blob Storage. Use modules for the VNet, the VM/app tier, and the database. Write a README explaining every resource and the decisions behind each one.


Month 3: Containers and Kubernetes

Containers are how modern applications are packaged. Kubernetes is how they are run at scale. Azure Kubernetes Service (AKS) is Microsoft's managed Kubernetes offering.

Docker fundamentals

  • Writing efficient, multi-stage Dockerfiles
  • Understanding layers, image size, and build caching
  • Managing images with Azure Container Registry (ACR)
  • Docker Compose for local multi-service development

Azure Kubernetes Service

AKS ConceptWhat You Need to Know
Node poolsSystem vs user node pools, VM SKU selection, autoscaling
NetworkingAzure CNI vs kubenet, ingress with AGIC or NGINX
IdentityWorkload identity, managed identities, pod identity
StorageAzure Disks vs Azure Files for PersistentVolumes
MonitoringAzure Monitor for containers, Prometheus integration
SecurityRBAC, network policies, pod security standards

Helm charts: How to package, version, and deploy Kubernetes applications with Helm. How to manage environment-specific values.

Month 3 project: Containerised microservices on AKS

Build a two-service application (e.g., an API and a frontend), containerise both, push to ACR, and deploy to AKS with Helm. Configure an Ingress controller, set up horizontal pod autoscaling, and add basic Prometheus metrics. Provision the entire AKS cluster with Terraform.


Month 4: CI/CD and DevOps Practices

Shipping code manually is a liability. Month 4 is about automating the full software delivery lifecycle.

GitHub Actions for Azure

GitHub Actions is the most common CI/CD tool for Azure workloads outside of Microsoft's own ecosystem.

Build pipelines that:

  • Run unit and integration tests on every pull request
  • Build and push Docker images to ACR on merge to main
  • Deploy to a staging environment automatically
  • Require manual approval for production deployments
  • Scan for vulnerabilities with Trivy or Snyk before any image push

Azure DevOps

Microsoft's integrated platform for source control, CI/CD, and project management. Dominant in large enterprise Azure shops.

Key components:

  • Azure Repos (Git hosting)
  • Azure Pipelines (YAML-based CI/CD)
  • Azure Artifacts (package registry for NuGet, npm, Python packages)
  • Service connections for deploying to Azure subscriptions

GitOps with ArgoCD

GitOps is the practice of using Git as the single source of truth for both application code and infrastructure state. ArgoCD watches your Git repositories and automatically syncs the cluster to the desired state.

Month 4 project: Full CI/CD pipeline

Build an end-to-end pipeline for your Month 3 microservices:

  • PR triggers unit tests and security scans
  • Merge to main builds, tags, and pushes images to ACR
  • ArgoCD detects the new image tag and deploys to staging
  • Manual approval gate before production promotion
  • Rollback strategy documented and tested

Month 5: Production Engineering and Job Preparation

The final month bridges the gap between "it works on my cluster" and "I can operate this in production."

Observability

The standard Azure observability stack:

  • Azure Monitor + Log Analytics: Centralised logs and metrics for all Azure resources
  • Application Insights: APM for application-level telemetry (requests, dependencies, exceptions)
  • Prometheus + Grafana: Open-source metrics and dashboards, integrates with AKS
  • Alerts and action groups: Notify Slack, email, or PagerDuty on threshold breaches

Write runbooks for your alerts. An alert with no runbook is noise. An alert with a clear runbook is signal.

Security and Governance

  • Azure Key Vault: Store all secrets, certificates, and connection strings — nothing hardcoded
  • Microsoft Defender for Cloud: Security posture management and threat detection
  • Azure Policy: Enforce compliance (e.g., all resources must have specific tags, no public blob storage)
  • Private Endpoints: Keep database and storage traffic off the public internet

FinOps Basics

  • Enable Azure Cost Management
  • Tag everything: environment, team, service, cost-centre
  • Set budgets and alerts on resource groups
  • Right-size VMs using Azure Advisor recommendations

Month 5 project: Production-ready platform

Take everything from Months 1–4 and harden it:

  • Add Key Vault for all secrets (no secrets in environment variables)
  • Add Azure Policy to enforce tagging
  • Configure Defender for Cloud
  • Build a Grafana dashboard with SLO panels
  • Write a post-mortem on one thing that broke during the project
  • Document architecture with a proper diagram and ADR log

Interview Preparation

What Azure cloud engineering interviews test in 2026:

Interview TypeExample QuestionWhat They Test
System designDesign a multi-region Azure architecture for a SaaS appCan you reason about availability, cost, and complexity?
Trade-off analysisAzure Functions vs AKS for a new microservice?Do you understand the operational and cost trade-offs?
SecurityHow do you prevent secrets from leaking in a CI/CD pipeline?Do you understand zero-trust and secrets management?
ObservabilityThe API response time has doubled. Walk me through diagnosing it.Is your mental model of the system correct?
CostThis resource group costs £5k/month. Reduce it by 25%.Do you understand the financial impact of architecture?

Your 5-Month Timeline

MonthFocusKey Output
1Cloud fundamentals + Azure basicsThree-tier app deployed manually, then via CLI
2IaC with Terraform + BicepFull infrastructure automated, state in Blob Storage
3Docker + AKSContainerised microservices running on AKS with Helm
4GitHub Actions + Azure DevOps + ArgoCDFull CI/CD pipeline with GitOps and approval gates
5Observability + Security + FinOpsProduction-hardened platform with SLOs, Key Vault, Policy

What to Do After Month 5

  • Write about each project on LinkedIn or a blog — one post per month of the roadmap
  • Contribute to an open-source Azure tool (Azure provider for Terraform, Bicep community)
  • Join the Azure community on Twitter, Discord, and local meetups
  • Apply confidently — you have five portfolio projects that prove production-level thinking

The portfolio is the differentiator. Certifications say you passed a test. Projects say you built something real. Build real things.


Found this useful? Share it with someone starting their cloud engineering journey. Drop a comment with which month you are on.

Part of the Learning Guides series · daraoladapo.com