A pipeline is an automated, ordered sequence of steps that builds, tests and ships your code every time it changes. Each step runs only if the previous one succeeds, so a single commit can flow from source control to a live environment without manual handoffs.
How does a CI/CD pipeline work?
A pipeline is triggered by an event - most often a push or pull request to a branch. From there it moves through stages such as installing dependencies, compiling or bundling the application, running automated tests, producing an artifact, and finally deploying that artifact to an environment.
Stages run in a defined order and share a workspace. If any step fails - a failing test, a broken build - the pipeline stops and reports the failure, so broken code never reaches production. This fast, automatic feedback loop is what makes continuous integration and continuous delivery possible.
Why do pipelines matter?
Pipelines replace error-prone manual release checklists with a repeatable, version-controlled process. Every change is built and tested the same way, which means fewer "works on my machine" surprises and faster, more confident releases.
They also create an audit trail: you can see exactly which commit produced which artifact and when it was deployed, making rollbacks and debugging far easier.
Example
A typical web-app pipeline might be: checkout code → install dependencies → run unit tests → build a production bundle → publish the bundle as an artifact → deploy it to a staging sandbox for review → promote to production. Each arrow is a gate that must pass before the next step starts.
Frequently asked questions
What is the difference between CI and CD?
CI (continuous integration) is the build-and-test half of the pipeline that verifies every change. CD (continuous delivery or deployment) is the half that packages and ships the verified change to an environment. Both run inside the same pipeline.
What is a pipeline stage?
A stage is a logical group of one or more steps in a pipeline, such as "test" or "deploy". Stages run in order, and a later stage only starts when the earlier one succeeds.
Can a pipeline deploy to more than one environment?
Yes. A common pattern is to deploy automatically to a sandbox or staging environment, then promote the same artifact to production after approval or after tests pass.
Suggest a new word or an edit to an existing one. Every submission is reviewed before it goes live.