How to Become a Cloud Engineer in 2026 (Free Path Included)

A cloud engineer builds and maintains the infrastructure that runs modern software. Instead of racking physical servers, you provision compute and storage from a provider like AWS, Azure, or Google Cloud, wire up the networking, and keep it all running as traffic grows and shrinks. On a normal day that looks like a handful of things: standing up servers and databases from a config file instead of clicking through a console, making sure a service stays available when one machine dies, watching the cost dashboard so the monthly bill does not quietly triple, handling security and access, and setting up the pipelines that ship code out to all of it. A question that comes up constantly: how is this different from DevOps? The honest answer is that the two overlap a lot, and at a small company one person often does both. The rough split: cloud engineering leans toward infrastructure architecture, how the pieces fit together and scale on a given provider, while DevOps leans toward the delivery side, the pipelines and tooling that get developers' code into production. Learn either path and you will pick up most of the other along the way. If the delivery side is where your interest sits, our guide on /guides/how-to-become-a-devops-engineer walks that route from Linux through CI/CD.

You do not need a long list of prerequisites, but a few things make the whole path smoother, and skipping them makes everything above feel like magic you cannot debug. Linux command line comes first. Almost every server, container, and cloud instance you will touch runs Linux, and nearly every tool assumes you can move around a shell, read logs, and manage files and permissions. You do not need to be an expert. A few hours of hands-on practice to get through the basics (moving between directories, editing files, running processes, reading output) covers most of what you need to begin, and you deepen it as you go. Basic networking is the second piece. You want a working mental model of IP addresses, DNS, and how an HTTP request travels from a browser to a server and back. This is not a full course; a few good free videos on YouTube get you to a usable level. It matters because cloud work is mostly connecting things together, and connections are networking. Programming is useful but not required to start. You can earn an entry cert and learn the core cloud services without writing much code. Once you reach infrastructure as code, though, you will want some scripting comfort (Bash or Python), because that layer is code. Do not let the lack of a programming background stop you from starting; add it when the path calls for it.

The skills stack, so the order matters. Here is the sequence that works, and why each step sits where it does. Step 1: Choose one cloud. AWS has the most job listings by a wide margin, so it is the default recommendation to start. Pick Azure if you are targeting enterprise or Microsoft-stack companies, common in government and large corporations, and Google Cloud if you are drawn to data-heavy and machine learning roles. Do not try to learn all three at once. The core concepts (compute, storage, networking, identity) transfer between them, so one deeply beats three shallowly. Step 2: Earn an entry-level certification. For AWS that is the Cloud Practitioner (CCP); for Azure it is AZ-900 Fundamentals. These are broad, vendor-friendly credentials that prove you understand how the cloud works, and they stay useful even if you switch providers later. This is your first concrete, resume-worthy target, and the free courses below line up with it directly. Step 3: Learn containers. A container packages an app with everything it needs to run, so it behaves the same on your laptop as in production. Start with Docker, since it is the container everyone uses, then move to Kubernetes, which runs and coordinates containers at scale. The order is not optional: Kubernetes only makes sense once containers feel obvious, so learners who jump straight to it end up memorizing commands they cannot reason about. Step 4: Learn infrastructure as code. This is what separates a cloud engineer from someone who clicks around a console. Terraform is the industry standard: you describe your infrastructure in files, check them into version control, and apply them to create or change real resources. Be warned, and we will be honest here: our catalog does not yet have a full free structured Terraform course. The official Terraform docs and HashiCorp's own tutorials are free and good, so that is where to go for this step for now. Step 5: Get an associate cert, then apply. Once the fundamentals are solid, the associate-level cert is the credential that actually opens interviews: AWS Solutions Architect Associate, or Azure Administrator AZ-104. Pair it with a project you deployed yourself, and you have enough to start applying for junior cloud roles.

You can cover most of this path without paying for the learning itself. Here is what to use for each stage and what each one is good for. AWS Cloud Practitioner Essentials (AWS Skill Builder, beginner, about 7 hours) is the official AWS training for the CCP exam, straight from Amazon and free. It is the cleanest structured overview of AWS core services, so it is where most people should begin the AWS route. Find it at /courses/aws-cloud-practitioner-essentials, with the platform profile at /platforms/aws-skill-builder. AWS Certified Cloud Practitioner Study Course (freeCodeCamp, beginner, 13 hours) is a single long YouTube video that walks the whole CCP exam. It is the best free option for dedicated exam prep because it is thorough and you can rewatch the parts that did not stick. Use it alongside the official essentials, not instead of it. See /courses/freecodecamp-aws-ccp and the platform at /platforms/freecodecamp. Microsoft Azure Fundamentals AZ-900 (Microsoft Learn, beginner, about 4 hours) is the Azure equivalent: free, official, and modular, so you can do it in short sittings. This is your entry cert prep if you pick Azure over AWS. It is at /courses/microsoft-learn-az-900, platform at /platforms/microsoft-learn. For containers, freeCodeCamp's Docker course (beginner, 2 hours) is a fast, practical intro at /courses/freecodecamp-docker, and its Kubernetes tutorial (intermediate) at /courses/freecodecamp-kubernetes builds straight on top of it. Do the Docker one first. AZ-400 Development for Enterprise DevOps (Microsoft Learn, intermediate, about 8 hours) at /courses/microsoft-learn-devops ties CI/CD, infrastructure as code, and monitoring together in an Azure context, and it is free with a certificate. Reach for it after the fundamentals. For the wider picture, our roundups at /guides/best-free-cloud-computing-courses-2026, /guides/best-free-aws-courses-2026, and /guides/best-free-devops-courses-2026 rank the free options in more detail.

Straight numbers, realistic rather than best-case. At about an hour a day, an entry cert (AWS Cloud Practitioner or AZ-900) is reachable in 6 to 8 weeks. An associate-level cert takes another 4 to 6 months on top of that, because it goes deeper and expects hands-on practice in a real account. Job-ready with a portfolio you can point to lands around 9 to 12 months for someone starting from outside tech. If you can study full-time, roughly halve those numbers. A developer already comfortable with Linux and how apps deploy can move faster still, since a good chunk of the foundation is already there. The figure that actually moves your timeline is not hours watched, it is hours building. An afternoon spent deploying something to a real cloud account and fixing what breaks teaches more than a week of video. Cloud providers offer free tiers precisely so you can practice on live infrastructure, so use one from the start rather than saving hands-on work for later.

Certifications carry more weight in cloud than in web development, and it is worth understanding why. A hiring manager cannot easily eyeball whether you understand cloud architecture, so an AWS or Azure cert acts as a baseline signal that you know the fundamentals. That is why the exam fee, unwelcome as it is on a free path, is usually worth paying for at least one cert when you are breaking in. Certs alone are not enough, though. Pair them with one real project. The strongest portfolio piece for a junior cloud role is a small app you deployed yourself on AWS or Azure, with the infrastructure defined in Terraform, a public GitHub repo showing the config, and basic monitoring wired up. That single project touches compute, networking, identity, infrastructure as code, and observability, which is a real slice of the actual job. Write a short README explaining your architecture and the tradeoffs you made; reviewers often read that first. On pay, entry-level cloud engineer roles in the US in 2026 tend to land in the $90k to $130k range, with wide variation by location and company. Remote and global rates swing further in both directions. If you are switching careers, the calm-under-pressure, incident-handling side of cloud work rewards people from operational backgrounds, so lean on that story. See /best-for/career-changers for how to frame it, and /learn/backend or /roadmap/backend if you want the application side to go with your infrastructure skills.