How to Become a Cloud Engineer in 2026: Skills, Tools & Portfolio
The WEF Future Jobs 2025 report projects 78 million net new jobs by 2030, with cloud-adjacent roles among the top 10 fastest-growing. Tech wage premiums for infrastructure roles have outpaced the national average since 2022 (verify current figures at bls.gov/eci).
What Does a Cloud Engineer Actually Do? (Day in the Life)
Cloud engineering isn't a single job description. It's a cluster of related roles unified by one trait: you own the infrastructure that everything else runs on. A morning shift might involve reviewing a Terraform PR that resizes an autoscaling group, then jumping into an incident where a misconfigured security group is blocking database connections, then ending with writing a runbook so the next on-call engineer doesn't have to reverse-engineer the fix.
Define AI roles and training. The AI Roles, Responsibilities & Training Policy: clarify who owns what for AI.
Your purchase helps keep our hubs free to read.
The specific title shapes the mix. A DevOps engineer spends more time on CI/CD pipelines and deployment automation. A site reliability engineer (SRE) tracks error budgets and writes postmortems. A cloud architect designs systems at the account and VPC level before anyone writes code. A platform engineer builds internal developer tooling: Kubernetes abstractions, golden paths, internal service catalogs. A cloud security engineer focuses on IAM policies, threat detection, and compliance frameworks.
What's consistent across all of them: you write code (IaC, scripts, sometimes application code), you read documentation constantly, and you own the reliability of systems you didn't build. On a bad week that means reverse-engineering a failing system at 2am from logs, runbooks, and guesswork.
The Technical Skills Stack: What You Need to Learn (and in What Order)
The mistake most beginners make is picking a cloud platform first and treating Linux or networking as optional. Cloud is abstracted infrastructure. Without the foundation, you can follow tutorials but you can't debug when they break.
Foundation Layer (Learn First)
- Linux CLI: File system navigation, permissions, processes, grep, awk, SSH, systemctl. You'll use this every day.
- Networking basics: TCP/IP, DNS resolution, subnets and CIDR notation, load balancers, firewall rules. Understanding cloud networking is where most beginners get stuck. Read it before you touch VPCs.
- Git: Not just commits. Branching strategies, pull requests, merge vs rebase, conflict resolution. Infrastructure code lives in Git.
- Security model: Identity and access management is the layer everything else depends on. IAM fundamentals apply across all three major clouds.
Cloud Platform Layer (One Provider, Then Expand)
Pick one. Trying to learn AWS, Azure, and GCP simultaneously is the second most common mistake. The concepts transfer between providers; the services don't. Pick AWS if your target employers lean enterprise US, Azure if they're Microsoft-heavy, GCP if they're data or AI-focused.
- AWS: Largest market share, most job postings, deepest service breadth. EC2, S3, VPC, IAM, RDS, Lambda as the starting cluster.
- Azure: Strong in enterprise environments, especially those already using Microsoft 365 and Active Directory. Microsoft Entra ID (formerly Azure AD), VNets, App Service.
- GCP: Preferred in data engineering and ML/AI workloads. BigQuery, GKE, Vertex AI.
Automation and Orchestration Layer
- Terraform: Write infrastructure as code, plan changes before applying, manage state. See our Terraform beginner guide for setup to first deployment.
- Docker + Kubernetes: Containers are how modern applications deploy. Kubernetes is how they run at scale. Start with Docker, then move to Kubernetes fundamentals.
- Python or Bash: Scripting for automation, API calls, and tooling. Python is more transferable. Bash is faster for quick one-offs.
- CI/CD pipelines: GitHub Actions, GitLab CI, or Jenkins. Understanding how code moves from commit to production is table stakes.
Essential Cloud Tools: What Every Engineer Uses Daily
Job postings list tools as requirements, but they're not all equal weight. Here's the realistic stack you'll touch in most cloud engineering roles.
Infrastructure and Deployment
- Terraform: Plan, apply, destroy. HashiCorp changed Terraform's license from MPL 2.0 to BSL in August 2023, prompting the community to fork the project as OpenTofu (Linux Foundation, MPL 2.0). IBM later acquired HashiCorp in a $6.4B deal announced April 2024. Teams that need open-source assurance are evaluating both.
- AWS CLI / Azure CLI / gcloud: Command-line access to cloud APIs. Essential for scripting, debugging, and CI/CD pipelines. Learning the CLI teaches you what's under the console's surface.
- Docker: Build, tag, push container images. Run containers locally. Write Dockerfiles. This is entry-level knowledge now, not advanced.
- kubectl: Control plane access for Kubernetes clusters. get, describe, apply, logs. You'll use this on-call.
Monitoring and Observability
- CloudWatch / Azure Monitor / Cloud Logging: Native cloud monitoring. Set up dashboards, alarms, and log queries before considering third-party tools.
- Prometheus + Grafana: Open-source metrics and dashboards. Common in Kubernetes environments.
- Datadog or Splunk: Enterprise observability platforms. Knowing what problem they solve matters in interviews even if you haven't used them.
Version Control and CI/CD
- Git (GitHub / GitLab / Bitbucket): Infrastructure as code lives here. Trunk-based development, branch protections, PR reviews for infra changes.
- GitHub Actions / GitLab CI: Automate Terraform plans on PR, apply on merge, run security scans. Building at least one CI/CD pipeline from scratch is a portfolio requirement.
Certifications That Get You in the Door
Cloud certifications won't replace portfolio projects, but they do one thing portfolios can't: they tell a recruiter you passed a vendor-administered exam on a specific body of knowledge. That matters when your resume is screened by someone who can't evaluate GitHub repos.
Where to start: AWS Certified Cloud Practitioner (CLF-C02) at $100 covers vendor-agnostic cloud fundamentals plus AWS specifics, with a three-year renewal cycle. It's the lowest barrier to a recognized credential and sets up the next tier. See the full cloud certifications roadmap for a multi-step progression across AWS, Azure, and GCP.
Free Training Resources (Verify Current Free Tier at Official URLs)
- AWS Skill Builder (skillbuilder.aws): 500+ digital courses on a free tier. Verify current free tier availability before planning your study path, as it has changed over time.
- Microsoft Learn (learn.microsoft.com/training): Free learning paths for all Azure certifications. The AZ-900 learning path is one of the better free certification prep resources available.
- Google Cloud Skills Boost (cloudskillsboost.google): Free tier with limited hands-on labs; paid tiers for full lab access. The free tier covers most conceptual content for the Associate Cloud Engineer exam.
Certification fees for Azure and GCP vary; verify current pricing at the official exam pages before budgeting. Figures from study guides or third-party sites frequently lag behind vendor pricing changes.
Building a Cloud Portfolio That Employers Notice
Certifications tell employers you know the concepts. Portfolio projects tell them you can build. Three to five well-documented projects, each solving a real problem and deployed to a real cloud account, outperform a resume full of course completions.
Host everything on GitHub with clear READMEs, architecture diagrams, and cost estimates. A repo with a diagram and a "what this costs to run" section signals that you think like an engineer, not just a tutorial follower.
Your First Cloud Job: Where to Look and What to Expect
The title on your first cloud offer probably isn't "cloud engineer." It's more likely Cloud Support Engineer, Junior DevOps Engineer, Platform Engineer I, SRE Associate, or Infrastructure Engineer. Search for all of these, not just the canonical role title.
Where to Find Listings
- LinkedIn Jobs: Filter by "Entry Level" and keywords like "cloud engineer," "DevOps," "SRE," "infrastructure." Set job alerts, not just saved searches.
- AWS, Azure, and GCP hiring pages: Cloud vendors hire large numbers of technical support and solution architect roles that are excellent cloud entry points.
- MSP (Managed Service Provider) roles: MSPs expose you to 5-10 different cloud architectures in year one. The learning curve is steeper, but the breadth builds faster than most corporate roles.
- Remote-first companies: Geographic range is wider than it was pre-2020. Distributed teams are standard in cloud engineering.
What to Expect in Interviews
Entry-level cloud engineering interviews typically cover three areas: conceptual knowledge (networking, IAM, compute types), hands-on skills (walk me through how you'd set up a VPC, what does this Terraform output mean), and behavioral questions about debugging, incident response, and how quickly you self-teach between roles.
Studying cloud concepts without deploying anything leads to answers that collapse under follow-up questions. Interviewers can tell the difference between someone who built a Kubernetes cluster and someone who only read about one. The portfolio is the preparation.
- At least one cloud certification (CLF-C02 minimum)
- 3+ portfolio projects on GitHub with READMEs and architecture diagrams
- Can explain VPC, subnets, security groups, and IAM roles from memory
- Terraform: can write a resource block, explain state, and describe remote backends
- Docker: can write a Dockerfile, build an image, and run it locally
- LinkedIn profile updated with cloud skills, cert badge, and project descriptions
- Can write a Python or Bash script to automate a real task (file parsing, API call, or server check)
Salaries, Seniority, and What the Career Path Looks Like
Data caveat: "Cloud engineer" does not map to a single BLS occupational code. Infrastructure roles span multiple SOC codes including 15-1244 (Network and Computer Systems Administrators), 15-1212 (Information Security Analysts), and 15-1211 (Computer Systems Analysts). All salary figures below are framed as "reported median compensation" and must be verified at current sources before use in negotiations.
The BLS Occupational Employment and Wage Statistics for SOC 15-1244 (the closest proxy for cloud/infrastructure roles) reports a median annual wage of approximately $95,360 as of May 2024. Verify current figures at BLS.gov. Cloud specialist premiums on top of that base vary significantly by certification level, geographic market, and provider expertise. Tech wage growth since fall 2022 has outpaced the national average (verify current figures at bls.gov/eci).
For US-specific role compensation broken down by level and company, levels.fyi and LinkedIn Salary Insights provide more granular data than BLS aggregates. Geographic caveat: US salary data does not reflect markets in the UK, EU, India, or APAC. For international comparisons, LinkedIn Global Talent Insights reports regional variation.
Seniority Levels
- Entry (0-2 years): Following runbooks, building from existing templates, learning operational patterns. Support-heavy, good for exposure breadth.
- Mid (2-5 years): Owns specific infrastructure domains, writes IaC from scratch, participates in architecture reviews, on-call rotation.
- Senior (5+ years): Designs multi-account architectures, mentors junior engineers, leads incident response, owns reliability objectives.
- Principal / Staff: Cross-team infrastructure strategy, cost optimization at scale, vendor negotiations, platform choices.
Staying Current: How Cloud Engineers Keep Their Skills Sharp
Cloud platforms ship multiple major service updates every week. Nobody reads all of it. The sustainable approach is a small signal stack that surfaces what matters for your focus area, combined with hands-on practice to convert reading into retained knowledge.
Signal Stack Worth Building
- AWS re:Post / Azure Updates / GCP Release Notes: Subscribe to the release notes RSS feed for your primary platform. Scan the titles weekly. Read the full post for anything touching services you operate.
- Cloud Native Computing Foundation (CNCF): The CNCF landscape tracks open-source cloud-native tools. Their annual survey surfaces what's in production use versus what's hyped.
- AWS re:Invent and Google Cloud Next talks: The architecture pattern talks, not the product announcements. These show how cloud teams at scale solve real problems. Available on YouTube, free.
- Last Week in AWS or similar curated newsletters: Filtered summaries that cut the firehose to what actually matters. One weekly read beats scanning the full vendor release blog daily.
Hands-On Practice That Compounds
Learning stops compounding if it stays in tutorials. Build your own cloud homelab (free tier or a small monthly budget), break things deliberately, and write up what you learned. A personal blog or GitHub notes file that documents your debugging process is also a slow-burn portfolio signal that accumulates over time.
Cloud certifications renew every three years. Treat renewal cycles as forced full re-engagement with the domain, not just exam prep. The exam content tracks platform changes, so the renewal reading reveals what's shifted in the past three years.
Video Resources
Common Roadblocks and How to Clear Them
Fix: Build 2–3 GitHub projects with Terraform or CloudFormation, deployed to a real cloud account. Show the architecture diagram, the IaC code, and the monthly cost estimate. Recruiters check GitHub before calendars.
Fix: Pair each certification with one deployed project that uses the services it covers. For AWS SAA-C03, build a three-tier app with EC2, RDS, and an Application Load Balancer — then tear it down with Terraform to show cost discipline.
Fix: AWS Free Tier covers the core services (EC2 t2.micro 750 hours/month, S3 5 GB, RDS 750 hours) for 12 months. Run
terraform destroy after every session to eliminate idle charges. Azure offers $200 credit for 30 days; Google Cloud gives $300. LocalStack lets you test most AWS services offline at zero cost.
Fix: Start with AWS — it holds the largest market share and the widest job pool. Get CLF-C02 (foundational, $100) then SAA-C03 (associate, $150). Add Azure AZ-900 once you're hired if your employer runs Microsoft stacks. Multi-cloud fluency follows naturally after the first certification.
Fix: AWS Educate, Microsoft Azure for Students, and Google Career Certificates provide free or subsidized access. AWS Skill Builder includes free practice exams. Once employed, most cloud-using companies reimburse certification costs — verify this in the job listing or confirm in the interview.
Cloud Engineering, AWS, Azure, Google Cloud, Kubernetes, Terraform, Docker, and related marks are trademarks of their respective owners. Salary data sourced from BLS OES (bls.gov) and WEF Future Jobs 2025 Report. Verify all figures at official sources before use in negotiations. Tech Jacks Solutions is an independent editorial publisher.