Glossary · devops
CI/CD
Definition
CI/CD (Continuous Integration / Continuous Deployment) is the DevOps practice of automatically testing, building, and deploying code changes to production. Modern teams deploy 5-50+ times per week; the failure rate drops 80% vs manual processes.
Detailed explanation
CI (Continuous Integration): automated on every PR — lint + type check + unit test + integration test + build. Target pipeline duration <10 min. CD (Continuous Deployment): merge to main → auto staging → smoke test passes → production deploy.
2026 tools: GitHub Actions (most common, GitHub-native), GitLab CI (GitLab-native), CircleCI (premium), Jenkins (legacy + flexible), ArgoCD (Kubernetes-native), Vercel/Netlify (Next.js + frontend-native).
Deploy strategies: blue-green (zero downtime, two environments), canary (new version on 5% of traffic), feature flags (LaunchDarkly, GrowthBook — code live but feature toggleable), rolling update.
DORA metrics: deployment frequency, lead time, change failure rate, MTTR (mean time to recovery). High-performing teams: 5-50+ deploys/week, lead time <24h, failure rate <5%, MTTR <30 min.
Use cases
→Modern web/mobile/SaaS development
→Continuous delivery discipline
→Error rate reduction + fast rollback
→DevOps + DevSecOps automation
Pros
- +80% lower error rate (auto tests)
- +10-100x faster deploys
- +Human error → minimum
- +1-click rollback
Cons
- −Initial setup 1-2 weeks
- −Pipeline cost (GitHub Actions $0-20/month/repo)
- −Needs test coverage (otherwise fake safety)
- −Hard in complex monorepos
Related terms
Related services
Planning a project around CI/CD?
In a 30-minute discovery call we share a written architecture + cost + team recommendation tailored to your project.
Start a discovery call