CI/CD Overview

First PublishedFeb 17, 2026Last UpdatedMar 31, 2026ByAtif Alam

CI/CD (Continuous Integration / Continuous Delivery) is the practice of automating the build, test, and deployment of software so that every code change is validated quickly and shipped reliably.

CI vs CD vs CD

The three terms are often grouped together but mean different things:

Term	What It Does	When It Runs
Continuous Integration (CI)	Merge code frequently; build and test every change automatically	Every push / PR
Continuous Delivery (CD)	Keep the application in a deployable state; deploy to staging automatically, production on approval	After CI passes
Continuous Deployment (CD)	Deploy every passing change to production automatically — no manual gate	After CI passes

Continuous Integration          Continuous Delivery              Continuous Deployment
─────────────────────           ─────────────────────            ─────────────────────
 push ──► build ──► test         ──► deploy staging              ──► deploy staging
                                     ──► manual approval             ──► deploy production
                                     ──► deploy production              (fully automatic)

Most teams practice Continuous Integration + Continuous Delivery — automated builds and tests with a manual approval gate before production. Fully automated Continuous Deployment requires high confidence in tests and monitoring.

Why CI/CD Matters

Without CI/CD	With CI/CD
”It works on my machine”	Every change builds in a clean environment
Bugs found days or weeks later	Bugs caught in minutes
Manual, error-prone deployments	One-click or automatic deployments
Infrequent, risky releases	Small, frequent, low-risk releases
No one knows if the build is broken	Build status is always visible

DevOps culture and agile delivery

CI/CD is a technical practice; DevOps is a cultural one: break down walls between development and operations through shared goals, fast feedback, and psychological safety. Agile methods (short iterations, backlog, reviews) pair well with small, frequent releases—but only if pipelines are trustworthy and production is observable. This library’s CI/CD best practices and observability sections support that loop; your org adds ceremonies and roles.

Anatomy of a Pipeline

A CI/CD pipeline is a series of automated steps that run every time code changes:

Trigger            Build             Test              Security          Deploy
────────           ─────             ────              ────────          ──────
push to main  ──►  compile     ──►  unit tests   ──►  SAST/lint   ──►  staging
pull request       install deps     integration       dependency       ──► approval
schedule           build image      e2e               scan             ──► production
manual                                                SBOM

Stage	Purpose	Examples
Trigger	What starts the pipeline	Push, PR, tag, schedule, manual, API call
Build	Compile code, install dependencies, create artifacts	npm install && npm run build, docker build
Test	Validate correctness	Unit tests, integration tests, end-to-end tests
Security	Catch vulnerabilities early	SAST, dependency scanning, container scanning
Deploy	Ship to an environment	Deploy to staging, wait for approval, deploy to prod

CI/CD Tools Landscape

Tool	Vendor	Config Format	Hosted Runners	Self-Hosted	Strengths
GitHub Actions	GitHub	YAML (.github/workflows/)	Yes	Yes	Marketplace, tight GitHub integration, OIDC
GitLab CI/CD	GitLab	YAML (.gitlab-ci.yml)	Yes (shared)	Yes	All-in-one platform, review apps, Auto DevOps
Jenkins	Open source	Groovy (Jenkinsfile)	No	Yes	Extremely flexible, huge plugin ecosystem
CircleCI	CircleCI	YAML (.circleci/config.yml)	Yes	Yes	Fast builds, Docker-native, orbs
Azure Pipelines	Microsoft	YAML (azure-pipelines.yml)	Yes	Yes	Azure integration, multi-stage, templates
AWS CodePipeline	AWS	JSON/Console	Yes (CodeBuild)	No	Native AWS integration, CodeBuild/CodeDeploy
ArgoCD	CNCF	Declarative (K8s manifests)	No	Yes (K8s)	GitOps, Kubernetes-native continuous delivery
Tekton	CNCF	YAML (K8s CRDs)	No	Yes (K8s)	Kubernetes-native CI/CD, cloud-agnostic

Choosing a Tool

Scenario	Good Fit
Code lives on GitHub	GitHub Actions
Code lives on GitLab	GitLab CI/CD
Need maximum flexibility / legacy pipelines	Jenkins
AWS-native infrastructure	AWS CodePipeline + CodeBuild
Azure-native infrastructure	Azure Pipelines
Kubernetes deployments (GitOps)	ArgoCD or Flux
Multi-cloud / vendor-neutral	GitHub Actions, GitLab CI, or Tekton

Topics in This Section

Start with Pipeline Fundamentals for platform-agnostic concepts, then dive into the specific CI/CD platforms (GitHub Actions, GitLab CI), followed by deployment strategies, GitOps, and best practices.

• Git Essentials — Clone, branch, PR workflow, revert, and tags for infra repos.
• Pipeline Fundamentals — Stages, jobs, triggers, artifacts, caching, secrets, environments, and runners.
• GitHub Actions — Workflows, triggers, marketplace, reusable workflows, matrix builds, OIDC, and self-hosted runners.
• GitLab CI/CD — .gitlab-ci.yml, runners, DAG pipelines, includes, environments, review apps, and Auto DevOps.
• Deployment Strategies — Rolling, blue/green, canary, feature flags, and rollback patterns.
• GitOps — Git as the source of truth for deployments with ArgoCD and Flux.
• Best Practices — Pipeline design, security, testing strategy, branch strategies, and DORA metrics.
• Security Scanning (DevSecOps) — SAST, SCA, container scanning, IaC scanning, secret detection, DAST, and SBOM.
• Release Management — Semantic versioning, changelog automation, semantic-release, release-please, and package publishing.
• Jenkins — Jenkinsfile pipelines, agents, plugins, shared libraries, and migration guidance.
• Artifact Management — Container registries, image tagging, lifecycle policies, promotion, and vulnerability scanning.
• Go for Ops — Tiny CLI with Go: modules, build, cross-compile, and when to use Go vs Bash or Python.
• Compliance and Audit — Audit trails, separation of duties, signed commits, change management, and SOC 2/PCI/HIPAA mapping.

Cloud-Specific CI/CD

For CI/CD services tied to a specific cloud provider, see:

• CI/CD on AWS — CodeBuild, CodeDeploy, CodePipeline, and GitHub Actions with AWS.
• DevOps on Azure — Azure Pipelines, Azure Repos, and GitHub Actions with Azure.

Key Takeaways

• CI catches bugs early by building and testing every change automatically.
• CD (Delivery) keeps the app deployable; CD (Deployment) ships every passing change to production.
• A typical pipeline: trigger -> build -> test -> security scan -> deploy staging -> approval -> deploy production.
• GitHub Actions and GitLab CI are the two most popular Git-platform CI/CD systems — both covered in this section.
• Cloud-specific CI/CD (AWS CodePipeline, Azure Pipelines) is covered in their respective topic sections.