A cloud engineer at a mid-size SaaS company recently posted their salary progression: $52K as a sysadmin in 2021, $118K as a cloud infrastructure engineer by 2024. The jump wasn't random—it followed a deliberate cloud computing roadmap: Linux basics → networking → AWS associate cert → a real project on their portfolio. No bootcamp, no CS degree. The roadmap was the differentiator.
This guide lays out that same roadmap in concrete stages, covering what to learn, in what order, which certifications move the needle for employers, and where to specialize once you've got the foundation. If you already know you want to go into cloud and you're trying to figure out where to start—or where to go next—this is it.
Stage 0: What the Cloud Computing Roadmap Actually Looks Like
Before touching any course, it helps to see the full path so you know what you're orienting toward. Most people who successfully break into cloud go through roughly the same sequence:
- Foundations — Linux, networking, basic scripting
- Cloud fundamentals — One provider deeply (AWS, GCP, or Azure)
- First certification — Associate-level to prove baseline competency to employers
- Specialization — DevOps/SRE, security, data/ML, or architecture
- Professional cert + portfolio project — The combo that clears hiring screens
The typical timeline for someone studying 10-15 hours per week: 9-15 months to the first cloud job. People with existing IT backgrounds (sysadmin, network admin) often compress this to 4-6 months. People starting from zero sometimes need 18+ months. Neither pace is wrong—the roadmap doesn't change based on how fast you run it.
Stage 1: Foundations Before You Touch a Cloud Console
Skipping this stage is the most common reason people stall out mid-roadmap. Cloud infrastructure is fundamentally distributed computing running on top of Linux and TCP/IP. If those are black boxes to you, every cloud concept will feel abstract and confusing.
Linux Command Line
You don't need to become a Linux sysadmin. You need to be comfortable enough to navigate a filesystem, read logs, edit config files with vim or nano, manage file permissions, and write a basic bash script. Most cloud CLI work happens in terminal sessions against Linux VMs. Spend 20-30 hours here. The Linux Foundation's free Introduction to Linux on edX is a reasonable starting point.
Networking Basics
VPCs, subnets, routing tables, security groups, load balancers—cloud networking is just networking with different UI. You need to understand IP addressing (CIDR notation), DNS, HTTP/HTTPS, firewalls, and what NAT does. You don't need to pass a CCNA, but you do need to understand why a web server in a private subnet can't be reached from the internet without a load balancer or NAT gateway in a public subnet.
One Programming Language at a Basics Level
Python is the right answer here. Not because you'll write application code, but because you'll write automation scripts, Lambda functions, cloud deployment scripts, and Infrastructure-as-Code. Enough Python to read and modify existing scripts, work with JSON/YAML, and call an API with the requests library. That's probably 20-40 hours of practice.
Stage 2: Pick One Cloud Provider and Go Deep on the Cloud Computing Roadmap
AWS has the largest market share (~32%) and the most job postings. GCP pays slightly higher on average and is dominant in data/ML workloads. Azure is Microsoft's ecosystem and owns enterprise accounts heavily tied to Office 365. Pick based on where you want to work, not based on which one is "better."
For most people entering the market in 2026, AWS is still the safest bet for job volume. GCP is the right call if you're targeting data engineering, ML infrastructure, or companies with heavy Google Workspace adoption.
What "Going Deep" Means
It means actually using the console, not just watching videos. Spin up a free-tier account. Deploy a VM. Set up a VPC with public and private subnets. Put a web app behind a load balancer. Store files in object storage. Set up IAM policies. Break things. Fix them. The hands-on hours matter more than the lecture hours at this stage.
Stage 3: Certifications That Actually Matter to Employers
Not all cloud certs carry equal weight. Here's what hiring managers actually filter on:
Tier 1 — Entry Certs (0-3 months in)
- AWS Cloud Practitioner / Google Cloud Digital Leader / Azure Fundamentals (AZ-900) — These prove you've oriented toward cloud but don't carry much weight alone. Get one, then move on quickly.
Tier 2 — Associate Certs (3-9 months in)
- AWS Solutions Architect Associate (SAA-C03) — The single most recognized cloud cert in job postings. If you're going AWS, this is the first meaningful milestone.
- AWS SysOps Administrator Associate — Better than SAA if you're targeting ops/SRE roles.
- Google Associate Cloud Engineer — GCP's equivalent. Pairs well with the foundational Google Cloud infrastructure courses.
Tier 3 — Professional Certs (12+ months in)
- AWS Solutions Architect Professional — Opens senior-level roles. Genuinely hard; don't rush it.
- Google Cloud Professional Cloud Architect — Equivalent weight on the GCP side. Required or strongly preferred at many enterprises using GCP.
- CKA (Certified Kubernetes Administrator) — Platform-agnostic, highly valued for DevOps/SRE roles across all three clouds.
The pattern that works: one associate cert → one specialization cert → portfolio project demonstrating what you built. That combination reliably clears HR screens.
Top Courses for the Google Cloud Path
If you're following the GCP track, these courses map directly to the Associate Cloud Engineer and Professional Cloud Architect roadmap. They're structured as Coursera specializations, meaning they build on each other—do them in order.
Essential Google Cloud Infrastructure: Foundation Course
The starting point for the GCP infrastructure track. Covers compute, storage, and the core networking primitives you'll use on every GCP project—exactly what you need before tackling more advanced topics like autoscaling or multi-region deployments.
Networking in Google Cloud: Fundamentals Course
VPCs, firewall rules, shared VPC, VPN, Cloud Interconnect—this covers the networking layer that underpins everything else in GCP. Worth doing before any cert prep because the cert exams lean heavily on networking scenarios.
Networking in Google Cloud: Routing and Addressing Course
Goes deeper on IP addressing, route management, and hybrid connectivity. Pair this with the Fundamentals course above if you're targeting a role that involves any infrastructure design work.
Elastic Google Cloud Infrastructure: Scaling and Automation Course
Covers autoscaling, load balancing, and infrastructure automation—the operational skills that separate engineers who can build things from engineers who can operate them at scale. Directly relevant to SRE and DevOps roles.
Managing Security in Google Cloud Course
IAM, organization policies, VPC Service Controls, security command center. Cloud security is one of the highest-paying specializations and consistently underrepresented among candidates—this course covers the foundational concepts for the GCP security track.
Modernize Infrastructure and Applications with Google Cloud Course
Covers containerization, Kubernetes on GKE, serverless with Cloud Run, and migration strategies. This is the course for people who want to move from infrastructure-only roles into platform engineering or application modernization work.
Stage 4: Specialization Paths and What They Pay
The cloud computing roadmap branches at the specialization stage. Here are the main tracks and realistic salary ranges for 2026:
Cloud/Infrastructure Engineer
Builds and maintains cloud environments. Terraform, IaC, CI/CD pipelines, VPC design, cost optimization. Median US salary: $115K-$140K. Most job postings. Lowest barrier to entry from a sysadmin background.
DevOps / SRE
Reliability, automation, and deployment pipelines. Kubernetes, Helm, ArgoCD, monitoring, incident response. Median US salary: $125K-$155K. High demand. Requires stronger programming fundamentals than pure infra work.
Cloud Security Engineer
IAM architecture, compliance, threat detection, security posture management. Median US salary: $135K-$165K. Consistently underfilled—there are fewer candidates than roles. The Google Cloud IAM and Networking for AWS Professionals course is specifically useful here if you're bridging from AWS to GCP security work.
Cloud Data / ML Engineer
Data pipelines, BigQuery, Spark on Dataproc, ML model deployment. Median US salary: $130K-$160K. GCP's dominance in this space makes the GCP data track particularly valuable. Requires stronger Python and SQL foundations.
Cloud Architect
Solution design, multi-cloud strategy, enterprise migration planning. Median US salary: $145K-$180K. This is generally where people land after 4-6 years of hands-on cloud work, not a direct entry point.
FAQ
How long does it take to follow a cloud computing roadmap from zero to employed?
For someone with no IT background, 12-18 months of consistent study (10-15 hrs/week) is a realistic range to reach the first junior cloud role. People with sysadmin, networking, or developer backgrounds often compress this to 4-8 months. The variable isn't intelligence—it's hours invested and whether you're building real things versus just watching videos.
Do I need a computer science degree to follow this roadmap?
No. Cloud roles are one of the more credential-agnostic areas of tech. Certifications function as a proxy for degree requirements at most companies. That said, a CS background does give you a leg up on the theory side (OS concepts, networking protocols, distributed systems). You can acquire those concepts without a degree, but it takes more deliberate effort.
Should I learn AWS, Azure, or GCP first?
Pick based on the job market in your target industry. AWS has the most job postings overall. Azure dominates enterprise/government sectors. GCP is strongest in data/ML and at companies using Google Workspace. If you have no strong industry preference, AWS is still the safest default for job volume.
Are cloud certifications worth it in 2026?
Associate-level certs (AWS SAA, Google ACE) still carry real weight at the hiring screen stage, especially for candidates without prior cloud job titles on their resume. Professional-level certs matter more in enterprise environments and consulting firms. The diminishing return kicks in when you're accumulating certs beyond your specialization—a portfolio project demonstrating real work typically does more than a third cert.
What's the difference between a cloud engineer and a cloud architect?
Engineers build and operate systems. Architects design them and make the technology decisions. In practice, the line blurs at smaller companies. Architect roles typically require 5+ years of cloud engineering experience and usually carry a $20K-$40K salary premium. Most people reach architecture from an engineering background, not directly.
Can I do this roadmap part-time while working a full-time job?
Yes, and most people do. The realistic pace at 10 hrs/week is slower but sustainable. The biggest risk is losing momentum between study sessions. Structured courses with deadlines (Coursera's subscription model applies light time pressure) tend to work better than self-paced video libraries where it's easy to drift.
Where to Go from Here
The cloud computing roadmap isn't complicated—it's sequential. Foundations first, then one cloud provider deeply, then your first associate cert, then specialize. The people who get stuck usually try to skip stages (jumping straight to Kubernetes without understanding basic networking) or study in circles without building anything concrete.
If you're starting today: spend the first month on Linux and networking basics, create a free-tier cloud account on your chosen provider, and pick one of the structured specialization courses above to work through systematically. The certification can wait until you've actually used the services. Build something, break it, understand why it broke. That's the part of the roadmap the courses can't give you—but they can give you everything else.