How to Become a Cloud Engineer in 2026 (Step-by-Step)

You can become a cloud engineer in 4-6 months with no prior tech experience. No degree is required. No certifications are required to start. What you need is a structured learning path, consistent practice, and a portfolio of projects that prove you can do the work.

This guide gives you the complete step-by-step roadmap: what cloud engineers actually do, the skills you need, the exact phases of learning, the certifications worth getting, and how to land your first role.

What cloud engineers do

Cloud engineers design, build, and manage the infrastructure that applications run on. That infrastructure lives in the cloud on platforms like AWS, Azure, or Google Cloud rather than in physical data centres managed by the company.

In practical terms, here's what a typical day involves:

Morning: Check monitoring dashboards. Review overnight alerts. Investigate any anomalies in application performance or infrastructure costs. Clear non-critical alerts and escalate anything that needs attention.

Mid-morning: Work on infrastructure projects. This might mean writing Terraform code to provision new AWS resources, building a CI/CD pipeline for a development team, or configuring Kubernetes deployments for a new microservice.

Afternoon: Code reviews on other engineers' infrastructure changes. Troubleshoot a deployment issue reported by a development team. Update documentation for a new process. Attend a sprint planning meeting where infrastructure requirements for upcoming features are discussed.

Late afternoon: Write a Python script to automate a manual process that's been taking 30 minutes each week. Push the code, open a pull request, and move on to planning tomorrow's work.

The work is a mix of building new systems, maintaining existing ones, automating manual processes, and collaborating with development teams. It requires both technical depth and communication skills.

Cloud engineering overlaps significantly with DevOps. In many companies, the titles are interchangeable. The distinction, where it exists, is that cloud engineers tend to focus more on infrastructure architecture and cloud platform services, while DevOps engineers focus more on CI/CD pipelines and deployment automation. Both roles use the same core toolset.

The step-by-step roadmap

The learning path has five phases. Each builds on the previous one. Skipping phases creates gaps that will slow you down later.

Phase 1: Foundations (4-5 weeks)

Linux, networking, Git, and Python.

Everything in cloud engineering runs on Linux. Every container, every server, every CI/CD runner. You need to be comfortable navigating the filesystem, managing processes, reading logs, and writing shell scripts.

Networking is the second foundation. Cloud infrastructure is fundamentally about networking routing traffic between services, configuring firewalls, setting up load balancers. You need to understand TCP/IP, DNS, HTTP, subnets, and how data flows between machines.

Git is how all infrastructure code is managed. Every Terraform file, every Kubernetes manifest, every CI/CD pipeline definition lives in a Git repository. Learn branching, merging, pull requests, and collaborative workflows.

Python is the automation language of cloud engineering. You'll use it to write scripts that interact with cloud APIs, process data, automate repetitive tasks, and build internal tools. Start with fundamentals: variables, functions, loops, file handling, and working with APIs.

What you should be able to do after Phase 1:

• Navigate a Linux server, manage files, read logs, write Bash scripts
• Explain how DNS resolution works and what happens when you type a URL into a browser
• Use Git for version control with branching and pull requests
• Write Python scripts that call APIs and process JSON data

Phase 2: Containers and CI/CD (3-4 weeks)

Docker and GitHub Actions.

Docker is how modern applications are packaged and deployed. You'll learn to write Dockerfiles, build container images, manage multi-container applications with Docker Compose, and understand container networking and volumes.

CI/CD (Continuous Integration / Continuous Deployment) is the automation backbone. Every code change triggers a pipeline that builds, tests, and deploys the application automatically. GitHub Actions is the most accessible place to start it's free, well-documented, and widely used.

What you should be able to do after Phase 2:

• Write a Dockerfile for a web application and build a production-ready image
• Create a Docker Compose file for a multi-container application (e.g., web app + database)
• Build a GitHub Actions pipeline that runs tests, builds a Docker image, and pushes it to a registry
• Explain the difference between CI and CD, and why automated pipelines matter

Phase 3: Cloud platform (3-4 weeks)

AWS core services.

AWS dominates the cloud market with roughly 31% market share. It's the most requested platform in job postings and the best place to build your cloud skills.

Focus on these core services first:

• EC2 virtual servers (compute)
• VPC virtual networking (subnets, security groups, route tables)
• S3 object storage
• IAM identity and access management (users, roles, policies)
• RDS managed relational databases
• Route 53 DNS management
• ELB/ALB load balancing
• CloudWatch monitoring and logging

Don't try to learn all 200+ AWS services. Learn these eight deeply. They cover 80% of what you'll use in your first role.

What you should be able to do after Phase 3:

• Deploy a web application on EC2 behind a load balancer
• Configure a VPC with public and private subnets, security groups, and routing
• Set up IAM users, roles, and policies following least-privilege principles
• Use S3 for static hosting and application storage
• Monitor infrastructure with CloudWatch dashboards and alerts

Phase 4: Infrastructure as Code and orchestration (3-4 weeks)

Terraform and Kubernetes.

Terraform lets you define cloud infrastructure in code files rather than clicking through the AWS console. Your entire infrastructure becomes version-controlled, reviewable, and reproducible. This is one of the most in-demand skills in cloud engineering job postings.

Kubernetes is the industry standard for running containerised applications at scale. It handles scheduling, scaling, networking, storage, and self-healing for your containers. It's complex, but understanding the fundamentals pods, deployments, services, ingress, and namespaces is essential.

What you should be able to do after Phase 4:

• Write Terraform configurations to provision AWS resources (VPC, EC2, RDS, S3)
• Use Terraform modules to organise reusable infrastructure code
• Deploy a containerised application to a Kubernetes cluster
• Configure Kubernetes services, ingress, and horizontal pod autoscaling
• Manage Terraform state and understand remote state backends

Phase 5: Advanced skills and capstone (2-3 weeks)

Monitoring, security, and a capstone project.

Monitoring and observability: set up Prometheus to collect metrics, Grafana to visualise them, and alerting rules to notify you when things go wrong. Understand the three pillars of observability metrics, logs, and traces.

Security: learn cloud security fundamentals including IAM best practices, network security (security groups, NACLs), secrets management, container image scanning, and basic compliance concepts.

Capstone project: build a complete end-to-end project that combines everything you've learnt. This becomes the centrepiece of your portfolio. A strong capstone project is worth more than any certification in a job interview.

What you should be able to do after Phase 5:

• Set up a monitoring stack (Prometheus + Grafana) for a Kubernetes cluster
• Implement security best practices across your infrastructure
• Build and deploy a complete cloud project from scratch infrastructure, CI/CD, monitoring, and documentation

Timeline: hours per week vs time to job-ready

Your timeline depends on how many hours per week you can dedicate:

Hours per week	Time to job-ready	Notes
10 hours	7-9 months	Steady but slow. Works alongside a full-time job.
15 hours	5-6 months	Good pace. Evenings + one weekend day.
20 hours	4-5 months	Recommended. Consistent progress without burnout.
30 hours	3-4 months	Intensive. Part-time work or between roles.
40+ hours	2-3 months	Full-time study. Fast but risk of burnout.

The 15-20 hours per week range is the sweet spot for most people. It's sustainable, allows time for concepts to settle, and leaves room for life outside of studying.

Consistency matters more than intensity. Studying 15 hours every week for 5 months produces better results than studying 40 hours per week for 3 weeks and then stopping.

Skills you need (priority ranked)

Not all skills are equally important. Here's the priority ranking for getting your first cloud engineering role:

Must-have (required for entry-level)

1. Linux command line, file management, permissions, processes, package management
2. Networking TCP/IP, DNS, HTTP, subnets, firewalls, load balancing
3. AWS core services EC2, VPC, S3, IAM, RDS (at minimum)
4. Docker Dockerfile, images, containers, Docker Compose
5. Git branching, merging, pull requests, collaborative workflows
6. CI/CD at least one pipeline tool (GitHub Actions recommended)
7. Terraform resource definitions, state management, basic modules
8. Python scripting, API interaction, automation

Should-have (strengthens your application)

9. Kubernetes pods, deployments, services, basic troubleshooting
10. Monitoring Prometheus, Grafana, or CloudWatch
11. Bash scripting beyond basic commands, actual automation scripts
12. Infrastructure design ability to diagram and explain architectures

Nice-to-have (differentiators)

13. Security fundamentals IAM best practices, network security, secrets management
14. Cost optimisation understanding of cloud pricing and cost reduction
15. A second cloud platform basic familiarity with Azure or GCP

Focus your energy on the must-haves first. A strong foundation in the top 8 skills will qualify you for the vast majority of entry-level cloud and DevOps roles.

Certifications worth getting

Certifications validate your knowledge and help you pass resume screening. They are not a substitute for hands-on skills, but they complement a strong portfolio.

AWS Solutions Architect Associate

The most recognised cloud certification. It tests your ability to design resilient, cost-optimised architectures on AWS. Covers compute, storage, networking, databases, security, and high availability.

When to get it: After completing Phase 3 (AWS core) or after your first few months in a role. Study time: 4-6 weeks alongside practical experience. Cost: $150 USD exam fee. Value: High. Widely recognised, frequently listed in job requirements.

HashiCorp Terraform Associate

Validates your infrastructure as code skills. Tests Terraform workflow, state management, modules, and HCL syntax.

When to get it: After completing Phase 4 (Terraform). Study time: 2-4 weeks. Cost: $70 USD exam fee. Value: High and rising. Terraform is the dominant IaC tool and this certification is increasingly requested.

CKA (Certified Kubernetes Administrator)

A hands-on exam where you solve real Kubernetes problems in a live environment. Tests cluster management, networking, storage, security, and troubleshooting.

When to get it: After 6-12 months of Kubernetes experience (on the job or through projects). Study time: 4-8 weeks of focused preparation. Cost: $395 USD (includes one free retake). Value: Very high for mid-level roles. The hands-on format makes it one of the most respected certifications in the field.

The strategy

Get your first role based on portfolio projects and interview skills. Then pursue certifications to accelerate career progression. The ideal order: AWS Solutions Architect Associate first, Terraform Associate second, CKA when you're ready for mid-level roles.

One portfolio project that deploys a real application on AWS with Terraform and CI/CD will outperform three certifications with no practical work. Certifications open doors. Portfolios close deals.

Building your portfolio

Your portfolio is the single most important asset in your job search. It proves you can do the work, not just talk about it. Aim for 3-5 projects that demonstrate increasing complexity.

Project 1: Static website on AWS with CI/CD

Deploy a static website to S3 with CloudFront (CDN). Set up a GitHub Actions pipeline that automatically deploys changes when you push to the main branch. Add Route 53 for a custom domain.

Skills demonstrated: AWS (S3, CloudFront, Route 53), CI/CD (GitHub Actions), Git.

Project 2: Containerised web application

Take a web application (Node.js or Python), write a Dockerfile, create a Docker Compose setup with a database, and deploy it to EC2 using a CI/CD pipeline. Add basic monitoring with CloudWatch.

Skills demonstrated: Docker, Docker Compose, EC2, CI/CD, basic monitoring.

Project 3: Infrastructure as Code with Terraform

Define a complete AWS environment in Terraform: VPC with public and private subnets, EC2 instances, an RDS database, security groups, and an application load balancer. Use Terraform modules to organise the code. Store state in an S3 backend.

Skills demonstrated: Terraform, AWS networking, infrastructure design, state management.

Project 4: Kubernetes deployment

Deploy a multi-tier application (frontend, backend, database) to a Kubernetes cluster. Configure services, ingress, horizontal pod autoscaling, and resource limits. Set up Prometheus and Grafana for monitoring.

Skills demonstrated: Kubernetes, monitoring (Prometheus + Grafana), container orchestration.

Project 5: Capstone full end-to-end platform

Combine everything: a containerised application deployed to Kubernetes on AWS, provisioned entirely with Terraform, with a full CI/CD pipeline, monitoring, alerting, and security scanning. Document the architecture with a clear README and architecture diagram.

Skills demonstrated: Everything. This is your showcase project.

For each project, create a public GitHub repository with:

• Clean, well-commented code
• A detailed README explaining the architecture, setup instructions, and what you learnt
• An architecture diagram (use draw.io or Excalidraw)
• Screenshots of the running application and monitoring dashboards

Employers will spend 2-3 minutes on your portfolio. Make those minutes count with clear documentation and visible results.

Getting your first job

Job titles to target

Don't limit your search to "Cloud Engineer." These titles all describe roles you're qualified for:

• Junior Cloud Engineer
• Junior DevOps Engineer
• Cloud Support Engineer
• Cloud Operations Engineer
• Infrastructure Engineer (Junior)
• DevOps Associate
• Platform Support Engineer
• Site Reliability Engineer (Junior/Associate)

Search for all of these. The same skill set applies across these titles. Companies use different naming conventions for essentially the same entry-level infrastructure role.

Where to find roles

• LinkedIn set up job alerts for each title above
• Indeed strong for UK-based roles
• Glassdoor useful for salary benchmarking
• Remote-specific boards We Work Remotely, RemoteOK, FlexJobs
• Company career pages target cloud-native companies, SaaS companies, and managed service providers (MSPs)

Interview preparation

Cloud engineering interviews typically have three components:

1. Technical screening (30-45 minutes). Questions about Linux, networking, AWS services, Docker, Terraform. "How does DNS work?" "What happens when an EC2 instance runs out of memory?" "Explain how a VPC is structured." Review fundamentals and be able to explain concepts clearly.

2. Practical exercise (60-90 minutes). Write a Dockerfile, create a Terraform configuration, troubleshoot a broken CI/CD pipeline, or debug a Kubernetes deployment. Your portfolio projects are your preparation if you've built the projects, you can handle these exercises.

3. Behavioural/culture fit (30-45 minutes). How you work in a team, how you handle incidents, how you prioritise tasks. Prepare specific examples from your projects: "When I was building my Terraform project, I encountered X issue and solved it by doing Y."

Tips that make a difference

• Tailor your CV to each role. Mirror the language from the job description. If they say "Infrastructure as Code," don't just write "Terraform" write "Infrastructure as Code (Terraform)."
• Lead with projects, not education. Your portfolio section should come before your education section.
• Quantify where possible. "Deployed a containerised application with 99.9% uptime" or "Reduced deployment time from 30 minutes to 3 minutes with CI/CD automation."
• Contribute to open source. Even small contributions (documentation fixes, bug reports) show you can work with real codebases and teams.
• Write about what you learn. A technical blog post about a problem you solved demonstrates communication skills and depth of understanding.

For more on breaking into the field with no experience, read our complete beginner's guide.

Career progression and salary expectations

Cloud engineering has a clear career ladder with strong salary growth at every level.

Junior Cloud Engineer / Cloud Support

UK salary: £40,000 £55,000 | US salary: $65,000 $90,000

Manage existing infrastructure, respond to alerts, deploy applications through established pipelines, write basic automation scripts. This is where you start. Typical time at this level: 1-2 years.

Mid-level Cloud Engineer / DevOps Engineer

UK salary: £55,000 £85,000 | US salary: $75,000 $140,000

Build CI/CD pipelines from scratch, write Terraform modules, manage Kubernetes clusters, design cloud architectures, automate complex workflows. You're building infrastructure, not just maintaining it. Typical time at this level: 2-3 years.

Senior Cloud Engineer / SRE / Platform Engineer

UK salary: £80,000 £120,000 | US salary: $120,000 $180,000

Design multi-region architectures, build internal developer platforms, lead incident response, mentor junior engineers, make technology strategy decisions. Three tracks diverge here: senior infrastructure, SRE (reliability focus), and platform engineering (developer experience focus). Typical time at this level: 2-4 years.

AI Infrastructure / MLOps Specialist

UK salary: £90,000 £140,000 | US salary: $130,000 $220,000+

The fastest-growing branch. Manage GPU clusters, build ML deployment pipelines, optimise inference costs. This specialisation layers AI-specific knowledge on top of strong cloud engineering fundamentals. See our AI infrastructure guide for details.

Cloud Architect / Infrastructure Lead

UK salary: £120,000 £200,000+ | US salary: $160,000 $300,000+

Set technical strategy, design enterprise-scale architectures, work with leadership on infrastructure budgets. This is the top of the technical ladder. 5-8+ years of experience.

The progression from junior to cloud architect typically takes 6-10 years. The jump in salary from entry level to senior is substantial you can realistically triple your starting salary within 5 years by consistently building skills and taking on more responsibility.

For a more detailed breakdown of every role at every level, see our cloud computing career guide.

Common mistakes to avoid

Mistake 1: Tutorial hell. Watching courses without building anything. After each phase, stop consuming content and build a project. The project is the learning.

Mistake 2: Certification stacking. Getting five certifications before applying for jobs. One or two certifications plus a strong portfolio is far more effective than five certifications with no projects.

Mistake 3: Learning too many tools at once. Docker, Kubernetes, Terraform, Ansible, Pulumi, CloudFormation, Jenkins, ArgoCD you don't need all of them. Learn Docker, Terraform, Kubernetes, and GitHub Actions. Add more tools after you're employed.

Mistake 4: Skipping networking fundamentals. Networking is boring. It's also the foundation of everything in the cloud. If you don't understand how traffic flows between services, you'll struggle with every other tool.

Mistake 5: Waiting until you feel "ready." You'll never feel 100% ready. Start applying for jobs when you've completed your portfolio projects. The interview process itself is a learning experience, and many people land roles before they feel fully prepared.

The bottom line

Cloud engineering is one of the most accessible and rewarding tech careers in 2026. The entry path is well-defined: Linux, networking, Python, Docker, CI/CD, AWS, Terraform, Kubernetes. The demand is real 28% year-over-year job growth with salaries starting at £40,000-55,000 (UK) or $65,000-90,000 (US).

You don't need a degree. You don't need prior tech experience. You need a structured learning path, consistent effort over 4-6 months, and a portfolio of projects that demonstrate what you can do.

The career ladder is clear and the ceiling is high. Junior to cloud architect. £40,000 to £200,000+. The work is challenging, the tools are constantly evolving, and the AI era is making every cloud engineering skill more valuable.

Start with Phase 1. Build your first project. Keep going.

Frequently Asked Questions

How do I become a cloud engineer with no experience?

What qualifications do you need to be a cloud engineer?

How long does it take to become a cloud engineer?

What is the difference between a cloud engineer and a DevOps engineer?

Is cloud engineering a good career in 2026?