If you are a software engineer moving into platform or DevOps work, the cert sequence below is the one that maps cleanly to the skills you will actually use. Skipping a step is fine if you have direct production experience in that area; the order matters because each cert assumes the prior layer.
The full sequence at a glance
- Linux foundation (LFCS or equivalent self-study)
- Docker and Kubernetes fundamentals (KCNA optional)
- CKA (Certified Kubernetes Administrator)
- One cloud associate cert (AWS SAA-C03 or AZ-104)
- HashiCorp Terraform Associate
- AWS DevOps Engineer Professional (DOP-C02) or Azure DevOps Engineer Expert (AZ-400)
Total time investment for a software engineer with no prior ops experience: 12-18 months of consistent evening study plus production work. For an engineer already shipping infrastructure code at work, 6-9 months.
Step 1: Linux foundation (4-6 weeks)
You cannot do platform engineering without comfort on the Linux command line. The bar: you can write a non-trivial bash script, debug a systemd service, read journalctl output, troubleshoot file permissions, configure SSH and sudo, and operate at the shell without StackOverflow open.
Cert option: Linux Foundation Certified System Administrator (LFCS), $395, performance-based 2-hour exam. Most software engineers can prepare in 4-6 weeks of evening work using the official Linux Foundation course material plus daily terminal practice on a Vagrant or Multipass VM.
Skip path: if you have used Linux daily for 2+ years (debugging build pipelines, running k8s containers, deploying servers), the cert adds little. Move on. The skills are the point.
Step 2: Docker and Kubernetes fundamentals (3-4 weeks)
Before CKA, you need conceptual fluency. The bar: you understand the difference between a container and a VM, you can write a useful Dockerfile, you understand Kubernetes pods, deployments, services, and namespaces, and you can read kubectl output without panic.
Cert option: KCNA (Kubernetes and Cloud Native Associate), $250, multiple choice. Optional but useful as a forcing function for structured study. Most engineers skip KCNA and go straight to CKA prep, using the first 3-4 weeks of CKA study as their fundamentals layer.
Free resource: KodeKloud Kubernetes for Beginners course is the most-recommended free option. Follow it with KodeKloud's CKA labs.
Step 3: CKA (Certified Kubernetes Administrator) (6-8 weeks)
The CKA is the cert that DevOps interviews actually probe for. It is hands-on (no multiple choice), 2 hours, performance-based on real Kubernetes clusters. Cost: $445, includes one free retake. The exam currently uses Kubernetes v1.34.
Why it matters: every modern platform team runs Kubernetes somewhere. CKA signals you can operate it, not just deploy YAML files.
Time investment: 6-8 weeks for someone with the prior layers solid. Daily lab work on Killer.sh (free with exam purchase) or a local kind/minikube cluster. The exam tests speed under pressure: 17 tasks in 120 minutes.
Step 4: one cloud associate cert (8-10 weeks)
Pick AWS or Azure based on what your employer runs or what your job market demands. Do not study both simultaneously.
AWS path: SAA-C03 (Solutions Architect Associate), $150. Tests architecture decisions across compute, storage, networking, identity. Most useful for DevOps engineers who will design infrastructure rather than only operate it.
Azure path: AZ-104 (Azure Administrator), $165. Tests operational configuration. More directly applicable to day-to-day DevOps tasks in an Azure environment.
Either choice is fine. The cloud-specific cert is what differentiates a Kubernetes-only operator from a platform engineer who can wire Kubernetes to managed cloud services (RDS, CloudWatch, IAM, KMS or their Azure equivalents).
Step 5: HashiCorp Terraform Associate (3-4 weeks)
HashiCorp Terraform Associate (003), $70.50, 60-minute multiple choice exam. The cheapest cert in this sequence and arguably the highest signal for DevOps roles. Terraform is the de facto IaC standard in 2026, and the cert tests fundamentals that every DevOps engineer is expected to know: state management, modules, providers, workspaces, lifecycle rules, the difference between count and for_each.
Why it matters: in DevOps job interviews in 2026, Terraform fluency is more probed than any single cloud cert. Holding the cert is a small signal; being able to write a working module from scratch is a large one.
Source: HashiCorp Certification page, May 2026. Renewal: 2 years, full re-sit at $70.50.
Step 6: capstone professional cert (10-14 weeks)
This is the cert that closes the loop and signals senior-level DevOps capability.
AWS path: DOP-C02 (DevOps Engineer Professional), $300, 180 minutes. Six domains covering SDLC automation, configuration management and IaC, monitoring and logging, incident response, security and compliance, and resilient cloud solutions. Difficulty: hard. Most candidates need 80-120 hours of study from an SAA-C03 or DVA-C02 baseline.
Azure path: AZ-400 (DevOps Engineer Expert), $165. Tests Azure DevOps Services and GitHub Actions integration with Azure, plus cross-cutting topics around secure pipelines, IaC, and monitoring. Easier than DOP-C02 in raw difficulty but more specific to Microsoft tooling.
Time investment summary
| Step | Cert | Cost | Time (zero baseline) | Time (engineer baseline) |
|---|---|---|---|---|
| 1 | LFCS (optional) | $395 | 4-6 weeks | 0-2 weeks |
| 2 | KCNA (optional) | $250 | 3-4 weeks | 1-2 weeks |
| 3 | CKA | $445 | 8-10 weeks | 4-6 weeks |
| 4 | SAA-C03 or AZ-104 | $150-165 | 8-10 weeks | 5-7 weeks |
| 5 | Terraform Associate | $70.50 | 3-4 weeks | 2-3 weeks |
| 6 | DOP-C02 or AZ-400 | $165-300 | 10-14 weeks | 6-10 weeks |
Total cost: roughly $1,200-$1,500 across all certs. Total time for a starting software engineer: 12-18 months. For an engineer already doing some infrastructure work: 6-9 months.
What this sequence signals on a resume
An engineer who completes this sequence has demonstrated: Linux fluency, container orchestration competence, hands-on cloud operations, IaC discipline, and full-stack pipeline design. That is the skill set hiring managers actually probe for in DevOps interviews. The certs are evidence; the work you do alongside them is what produces the offers.
Verify your shell setup before starting
# Confirm the basic toolchain a DevOps engineer needs locally
docker --version
kubectl version --client
terraform version
aws --version || az --version
git --version
If any of these commands return an error or "command not found," fix it before opening a study course. The roadmap above assumes you have a working local toolchain on day one.