Cloud & Infrastructure
Cloud and infrastructure engineering is the work that goes into building and maintaining the platforms that software runs on. Infrastructure engineers decide how applications are deployed, how they scale, how data is stored, and how the system recovers when something breaks. Experienced cloud engineers are in high demand and the gap between supply and demand has been growing for several years.
Why Linux is the starting point
Competence using the operating system Linux is essential for cloud and infrastructure engineers because most servers running on the internet run Linux. This includes cloud instances, containers and most CI/CD pipelines, so it’s essential to be able to navigate a Linux system from the command line. This is not a skill that’s ‘nice to have’ but can ultimately be worked around. It’s a foundational part of the work.
What to learn
First, install the free operating system Ubuntu on a virtual machine or use a cloud free tier to get a remote server. Then, start getting familiar with it: make sure you know how processes work, how the file system is structured, how permissions work, how to install software, and how to read logs. Learn how to connect to a remote server over Secure Shell (SSH). You’ll be able to learn most of this within a week, and it’s not optional, it’s essential.
The networking knowledge you actually need
Although you don’t need to be a network engineer, you do need to understand why a request from a browser reaches a server, what an IP address and a port are, what DNS does, and what happens when a firewall blocks traffic. These are the issues that come up the most frequently when trying to diagnose why something isn’t working as it should.
What breaks most often
The majority of “why is this not working?” infrastructure problems can be broken down into three areas: DNS, certificates, or firewall rules. Getting to know these three areas deeply will help you to resolve most of the incidents you face in your first few years. Network knowledge is also essential for understanding cloud security, including virtual private clouds, security groups, and subnets.
Which platform to start with
As the platform with the largest market share, AWS also has the most job postings. Azure is dominant in enterprises that already use Microsoft tools, while Google Cloud is strong in data and machine learning workloads. If neither of these are factors that you need to consider, then start with AWS. The concepts are often transferable between platforms, and the foundational certifications (AWS Cloud Practitioner, then Solutions Architect Associate) are widely recognised by employers.
Get hands-on
Do more than just read documentation. Create an account using the free tier (and set a reminder to cancel if you don’t want to pay once billing kicks in), then spin up a virtual machine, put a simple web server on it, point a domain at it and visit it in a browser. Then destroy it. Get your infrastructure knowledge to stick by building and breaking things, playing around and seeing how it works.
What a container is
In the context of software and cloud computing, a container is a lightweight, standalone, and executable package that contains everything needed to run an application (the code, the runtime, the libraries, and the configuration) and means that it runs the same way everywhere, regardless of what is on the host machine. This solves the “it works on my laptop” problem. Docker is the tool most commonly used to build and run containers, and it’ll be hard to understand modern infrastructure without being able to use it.
What to build
Most modern software deployment pipelines have three steps, which you can mimic in the process of your learning. Take a simple web application and write a Dockerfile that packages it into an image. Run it locally. Push the image to a container registry (Docker Hub has a free tier).
Why clicking in a console is not enough
Although manually creating cloud resources through a web interface is fine for learning, in production, it creates problems. Nobody knows exactly what was configured, and there is no history of changes. Recreating the environment in a different region requires doing everything by hand again. Infrastructure as Code (IaC) solves this by describing the desired infrastructure in files that can be version-controlled, reviewed, and applied automatically. This allows teams to treat infrastructure the way they would treat software: versioned, tested and deployed rapidly and reliably.
Terraform is the standard
Terraform is the most widely adopted IaC tool. You write configuration files that describe what cloud resources you want (a server, a database, a load balancer), run terraform apply, and it creates them for you. Learn to write basic Terraform, understand state management, and know what a plan output tells you before you apply it.
What Kubernetes does
Kubernetes (sometimes referred to as K8s) is a container orchestration platform used for running containers at scale across a cluster of machines. It handles deploying new versions of your application, scaling the number of running copies up and down, restarting containers that crash, and routing traffic between them. It is the standard for running containers in production at any meaningful scale.
Where to start without a large cluster
Run Kubernetes locally using minikube or kind, or use a managed service like Google Kubernetes Engine (which has a free tier). There are four core objects you should learn: a Pod (a running container), a Deployment (how you manage Pods), a Service (how traffic reaches your Pods), and a ConfigMap (how you inject configuration). Those four concepts cover most of what you encounter in practice.
