Cloud Computing & DevOps

1. What the cloud actually is

The cloud is other people's computers, rented by the second, with an API. This lesson separates the marketing from the engineering: the three service models and what each one moves to the provider, regions and availability zones, the shared-responsibility line that decides who fixes what, the core service categories you will use, and the pricing model that makes a badly designed system expensive while a well-designed one scales down as well as up.

2. Cloud-native architecture

The cloud rewards systems that assume machines can disappear and demand can change: stateless services, externalised state, horizontal scaling, and failure handled as a normal event rather than an exception. This lesson covers those patterns — load balancing, autoscaling, object storage, managed databases, queues, caching, serverless — and the discipline of designing for cost and resilience at the same time, because in the cloud the two are the same conversation.

1. Containers and infrastructure as code

Containers solved the oldest problem in delivery — it works on my machine — by packaging an application with its dependencies into one immutable image. Infrastructure as code solved the second — nobody remembers what we clicked — by describing servers, networks and databases in versioned files. This lesson covers what a container really is, how images and registries work, what orchestration adds, and how to write infrastructure that can be reviewed, reproduced and rolled back.

2. CI/CD, observability and incident response

Delivery is not a file transfer; it is a pipeline with tests, and a deploy you can undo. This lesson covers what belongs in a pipeline and why, the difference between continuous delivery and continuous deployment, strategies for releasing without downtime, and then the part most teams learn too late: observability — logs, metrics, traces and SLOs — plus an incident process that turns a bad hour into a permanent improvement.