Yes, you can become a cloud engineer with no IT background — and the honest roadmap takes about 9 to 12 months of steady part-time study, not the 30 days some courses promise. The path is: learn how computers and networks actually work, get hands-on with one cloud platform, earn one entry-level certification, then build three small projects that prove you can do the job. This guide lays out each milestone in order, with the free resources and portfolio projects that matter at every stage.
Most career-change guides skip the uncomfortable part, so let me say it first: cloud engineering is not a shortcut into tech. Employers hire cloud engineers to keep systems running, and they will test whether you can do that. The good news is that none of what they test requires a computer science degree. It requires a specific set of skills, all of them learnable in public, mostly for free — and hiring managers care far more about what you have built than what you studied.
What Does a Cloud Engineer Actually Do?
Before you commit a year to this, make sure the job is what you think it is. A cloud engineer designs, builds, and maintains the infrastructure that runs on cloud platforms like AWS, Azure, or Google Cloud. On a normal day that means launching and configuring virtual servers, setting up storage and databases, wiring networks together, automating repetitive tasks with scripts, and fixing things when they break at 2 a.m.
It is not the same as software development. You will write some code — mostly short scripts, not full applications — but the core skill is understanding systems: how a request travels from a user’s browser to a server, where it can fail, and how to make it reliable. If you enjoy taking things apart to see how they work, and you are comfortable with the idea that part of the job is detective work when something stops working, you will like it. If you wanted a job that never involves troubleshooting, look elsewhere — troubleshooting is the job.
The pay reflects that responsibility. Entry-level cloud roles in the US typically start around $70,000–$90,000, with mid-level engineers earning $110,000–$140,000. Those numbers vary by location and company, but the direction is consistent: it is one of the better-paid career changes available without a degree.
The Roadmap: Six Milestones, in Order
This roadmap is built for someone studying after work, roughly 10–12 hours a week. Each milestone ends with something you can show — a project, a certificate, or a working system — because employers hire proof, not intentions.
Milestone 1: Computer Basics (Weeks 1–4)
You cannot skip this, and it is where most career changers quietly fail. Cloud platforms assume you already know how computers work, and their tutorials will not slow down to explain it. Spend a month on the fundamentals: how operating systems work, what the command line is and why professionals live in it, how files, permissions, and processes fit together, and how the internet actually moves data — IP addresses, DNS, HTTP, and ports.
The practical version: install Linux (Ubuntu is the friendliest start) on an old laptop or a virtual machine, and force yourself to do everyday tasks from the terminal for a month. Learn the twenty commands that do 90% of the work — navigating directories, copying and moving files, checking what is running, reading logs. Free resources like Professor Messer’s CompTIA A+ videos or the “Linux Journey” site cover this ground well, and you do not need to pay for the A+ exam itself unless you want the credential. The goal is comfort, not a certificate.
Portfolio proof: none yet — this milestone is about fluency, and it shows up in everything after it.
Milestone 2: Networking Fundamentals (Weeks 5–8)
Networking is the subject career changers underestimate most, and the one interviewers probe hardest. A cloud engineer who does not understand networking is a liability, because nearly every cloud problem is a networking problem in disguise.
Learn subnets and CIDR notation until you can calculate them without a calculator, the difference between public and private IP addresses, what a VPC is and why every cloud workload lives inside one, and how firewalls and security groups decide what traffic gets through. The single most useful exercise here is drawing a network diagram of a simple system — a web server, a database, and the firewall rules between them — and then explaining out loud why each rule exists. If you can explain it, you understand it.
Milestone 3: Pick One Cloud and Go Deep (Weeks 9–20)
Pick AWS first. Not because it is the best platform, but because it has the most jobs, the most free learning material, and the most forgiving free tier. Azure is a strong second choice if you already work in a Microsoft-heavy environment; Google Cloud is a fine platform with fewer entry-level openings. The wrong move is trying to learn all three at once. Depth in one beats surface knowledge in three, every time.
Work through the free AWS Skill Builder courses and build as you learn. Launch a virtual server and host a simple website on it. Store files in S3 and serve them through a content delivery network. Set up a database, then break the connection on purpose and fix it. The goal is not to collect course-completion badges — it is to be able to do these things without following a tutorial step by step.
Portfolio proof: a live personal website hosted on AWS, with its architecture documented — what each service does, what it costs per month (aim for under $5), and what you would change if traffic grew tenfold. That one page tells a hiring manager more than a CV full of course names.
Milestone 4: Earn One Entry-Level Certification (Weeks 21–28)
Certifications do not make you a cloud engineer, but for a career changer they do something specific and valuable: they get your CV past the first filter. Hiring managers who have never met you need a reason to believe you, and an entry-level cert is that reason. The right first one is the AWS Certified Cloud Practitioner — it is the cheapest, the broadest, and the one most entry-level job adverts mention by name.
Study with the free AWS exam guide and one practice-test set (the official practice exam costs $20 and is worth it; skip the $100+ video courses, which mostly repackage the free material). Take practice tests until you consistently score above 85%, then book the exam. The Cloud Practitioner costs $100 and takes about 90 minutes. If you fail, you can retake it after two weeks — a failed first attempt is common and says nothing about your chances.
Milestone 5: Learn Infrastructure as Code (Weeks 29–36)
This is the milestone that separates candidates who get interviews from candidates who get offers. Modern cloud teams do not click through consoles to build things — they write code that builds things, so infrastructure can be versioned, reviewed, and rebuilt identically every time. The tool to learn is Terraform: it works across all three major clouds, its syntax is readable, and “Terraform” appears in more cloud job adverts than any other tool name.
Rebuild your Milestone 3 website using Terraform instead of clicking. When your infrastructure is a file you can delete and recreate with one command, you have crossed from “person who has used AWS” to “person who works the way real teams work.”
Portfolio proof: your project on GitHub, with the Terraform code, a README explaining the architecture, and a cost breakdown. Recruiters do check GitHub — a small, clean, documented repository beats an empty profile every time.
Milestone 6: Apply Like It Is a Job (Weeks 37+)
Start applying when you have the certification and two solid projects — not when you feel ready, because you will never feel ready. Target cloud support engineer, junior cloud engineer, and IT support roles at companies that use AWS. Support roles are the most realistic first step: they teach you how real systems break, they pay you to learn, and they are the most common route into cloud engineering for people without a tech background.
Tailor each application. Mention the specific cloud services the job advert names, and link to the project that uses them. In interviews, walk through your projects the way you would walk through a real system — what it does, what went wrong while you built it, and how you fixed it. “I broke the database connection and spent a day learning how security groups work” is a better interview answer than “I completed a course,” because it is a true story about solving a problem.
How Long Does It Really Take?
At 10–12 hours a week, the honest answer is 9–12 months from zero to first job. Some people do it faster — usually people who already work adjacent to technology, like IT support or data analysis. Some take 15 months, usually because life intervenes, which is fine. The number that matters is not months but consistency: two hours a day, five days a week, beats weekend binges every time, because cloud skills are built by repetition, not by cramming.
Do not believe any course that promises a cloud engineering job in 30 days. The certification alone takes a month of real study, and employers can tell the difference between someone who memorised exam answers and someone who has actually built things. The shortcut is a trap: it costs money, it does not work, and it leaves people discouraged enough to quit a path they could have completed.
The One Mistake That Costs Career Changers the Most
The most expensive mistake is collecting certificates instead of building things. I have seen CVs with four cloud certifications and zero projects, and they lose to candidates with one certification and a GitHub full of working systems. Certifications open doors; projects get you through them. If you are choosing between a second exam and a second project, build the project.
Here is a simple test to keep yourself honest: once a month, try to explain your current project to a non-technical friend in five minutes — what it does, why each part exists, and what it costs to run. If you cannot do that, you are studying too passively. Go back to the keyboard and build something.
Frequently Asked Questions
Can I become a cloud engineer with no IT background at 40?
Yes — age is not the barrier people fear it is. Employers care about whether you can do the work and learn quickly, and career changers often bring strengths fresh graduates lack: reliability, communication skills, and experience working with customers. The realistic path is the same at 40 as at 25, though you should target employers who value experience over pedigree, such as mid-sized companies and public-sector teams rather than the most competitive big-tech graduate schemes.
Do I need a computer science degree to become a cloud engineer?
No. Most cloud engineer job adverts list a degree as “preferred” rather than “required,” and many drop the requirement entirely. What employers actually test is hands-on ability: can you build a working system, secure it, and explain your choices? A degree helps with theory, but the practical skills — and the portfolio that proves them — matter more for getting hired.
How much does it cost to become a cloud engineer?
You can complete this entire roadmap for under $300: the Cloud Practitioner exam ($100), one practice exam ($20), and a few dollars a month of AWS free-tier usage. Everything else — Linux, Terraform, AWS Skill Builder, and the documentation — is free. Avoid bootcamps that charge thousands; the free material is the same content, and no bootcamp certificate carries more weight with employers than a project you built yourself.
Which cloud should I learn first — AWS, Azure, or GCP?
Learn AWS first unless you have a specific reason not to. It has the largest job market, the most free learning resources, and the most forgiving free tier for practice. If you already work in a company that runs on Microsoft, Azure is the smarter choice because you can practise on systems you already touch. Google Cloud is worth learning second, after you are employed — the concepts transfer, and a second cloud on your CV is a promotion argument, not a starting requirement.
