AWS certification path: the order to take them

The AWS certification levels

AWS organises its certifications into four tiers, and knowing what each one assumes stops you over- or under-shooting:

• Foundational — Cloud Practitioner and AI Practitioner. Concepts, core services, pricing and security basics (or, for AI Practitioner, AI and generative-AI fundamentals on AWS). No technical background assumed.
• Associate — Solutions Architect, Developer, CloudOps Engineer, Data Engineer and Machine Learning Engineer. The recognised professional entry tier; AWS recommends around a year of hands-on experience for each.
• Professional — Solutions Architect and DevOps Engineer. Deep, scenario-heavy exams that assume years of real experience.
• Specialty — narrow, deep exams (security, networking and others) for people already working in that niche.

There is no required order — which changes the question

AWS imposes no prerequisites on any certification exam. You do not need Cloud Practitioner to sit an associate, and you do not need an associate to sit a professional. Anyone can book any exam at any level, which means “what order must I follow?” is the wrong question. The real question is “what order gets me to my goal fastest without wasting fees?” — and that depends on two things: how much cloud experience you have today, and what work you want the certification to lead to.

The levels do still mean something, though, even without formal gates. Each tier assumes the knowledge of the one below it and recommends progressively more hands-on experience — the associates suggest around a year of AWS experience, the professionals more. Skipping levels is allowed; skipping the knowledge is not. An order that outruns your experience mostly buys retake fees, since every attempt is paid in full and a fail means a 14-day wait before rebooking.

The sensible default order

For most people the path resolves to three moves:

1. Start with Cloud Practitioner (CLF-C02) if you are new to the cloud. It is the only exam designed on the assumption of no hands-on experience, its syllabus doubles as a structured introduction to AWS, and passing earns a 50% discount voucher toward your next exam — so the foundational exam partly funds the associate that follows.
2. Take one associate matched to your target role. This is the credential hiring managers recognise fastest for cloud work, and choosing by role means the studying doubles as job preparation rather than trivia collection.
3. Leave professional and specialty exams until experience justifies them. They are written for people who already do the work at depth — advanced architecture, DevOps, security, networking — and attempting them early is the most expensive way to discover that.

The three general-purpose associate tracks

The associate tier is where the path forks by role, and this fork matters more than any question of ordering. Solutions Architect – Associate (SAA-C03) is the design track: choosing and combining services into resilient, cost-effective architectures. It is the broadest and most widely requested associate, which makes it the strong default when your direction is still forming. Developer – Associate (DVA-C02) is the building track: applications on AWS, SDKs and APIs, serverless, deployment — the natural fit if you write software for a living. CloudOps Engineer – Associate (SOA-C03, formerly SysOps Administrator) is the operations track: running, monitoring and troubleshooting workloads in production.

Choose by the job description you want to answer, not by reputation. If you would design systems, take Solutions Architect; if you would build them, Developer; if you would keep them running, CloudOps. The three share a large common core — accounts, networking, security, the major services — so a second associate later is cheaper to earn than the first. But start with the one whose syllabus reads like your target role, because that is the one whose preparation pays twice.

Which certification for which role

If you know the job you want, the path mostly picks itself. Map the role first, then work backwards to the exam:

• Cloud engineer, infrastructure or platform engineer — Solutions Architect – Associate. It is the broadest associate and the one named most often in cloud-engineering job descriptions.
• Software developer building on AWS — Developer – Associate. It goes deepest on serverless, APIs, SDKs, permissions and deployment tooling.
• Operations, SRE or systems administration — CloudOps Engineer – Associate. Monitoring, automation, reliability and running workloads in production; often the most natural step from a sysadmin background.
• Data engineering — Data Engineer – Associate. Ingestion, transformation, pipelines and data stores on AWS.
• Machine learning engineering — Machine Learning Engineer – Associate, usually after the AI Practitioner if generative AI and ML are new to you.
• Security — an associate first (most often Solutions Architect), then the Security – Specialty once security is your day job.
• Non-technical roles around cloud projects (sales, product, finance, management) — Cloud Practitioner, and often nothing further. That is a complete path, not an unfinished one.

When to skip Cloud Practitioner — and when not to

Engineers already working hands-on with AWS can reasonably skip straight to an associate. For someone who deploys to AWS weekly, the foundational syllabus is largely known material, and the associate is the better use of a fee and a study window. Nothing downstream requires the foundational credential, and no interviewer will ask why you do not hold it if you hold an associate.

Beginners and career changers should usually not skip it. The associates assume the foundational content — the service landscape, the shared responsibility model, pricing behaviour — and studying Cloud Practitioner first is effectively the opening chapter of associate preparation with a credential at the end of it. Non-engineering professionals around cloud projects — product, sales, finance, management — often stop at Cloud Practitioner entirely, and that is a perfectly complete path: it was designed for exactly that audience. Our guide on the easiest AWS certification covers this starting-point decision in more depth.

Sequencing the practicalities: vouchers, validity, renewals

A few mechanics reward a deliberate order. Every pass earns a 50% discount voucher toward your next exam, so chaining exams — foundational into associate, associate into a later professional — reduces the total spend compared with taking them as isolated events. Fees at the time of writing are $100 for Cloud Practitioner, $150 for the associates and $300 for professional and specialty exams; check AWS’s current pricing page, as these change.

Validity also shapes sensible pacing. Each certification lasts three years, and renewing by passing a higher-level exam also renews the lower-level ones — so a path that steps upward keeps the earlier credentials alive as a side effect. There is no need to rush the whole catalogue inside one validity window; a well-chosen associate backed by real projects does more for a career than a shelf of badges earned in a sprint. Order is a servant of the goal, and the goal is the role, not the collection.