Continuous integration merges every developer's work into a shared branch many times a day, and a server builds and tests each change at once. Continuous delivery goes further: every change that passes can be released safely, and continuous deployment releases it automatically. Together they turn releases from events into routine.
Key concepts
- Fast feedback: a pipeline that fails within minutes tells the author while the change is still fresh.
- Build once, promote: one artifact is built, then the same one moves through test, staging and production.
- Test stages: quick unit tests first, slower integration and end-to-end tests after.
- Deployment strategies: rolling, blue-green and canary releases limit the harm of a bad change.
- Pipeline as code: the pipeline is a file in the repository, reviewed like any other.
Vocabulary
- Artifact: the built output of a pipeline, such as a package or container image.
- Runner or agent: the machine that executes pipeline jobs.
- Flaky test: a test that sometimes fails without any code change; fix or remove it.
- Feature flag: a switch that turns code on or off at runtime, separating deploying from releasing.
What to study, in order
- Git branches and merges, and how a pull request is reviewed.
- Write a test suite for a small project that runs with one command.
- Make a pipeline that runs that command on every push and blocks a merge when it fails.
- Build an artifact in the pipeline and deploy it to a test environment automatically.
- Add a canary or blue-green release and a one-step rollback.
- Measure the pipeline's duration and shorten it with caching and parallel jobs.