AWS Certified Data Engineer – Associate
Data Engineer – Associate examines the plumbing: ingesting data, transforming it, storing it somewhere queryable, and keeping the whole arrangement governed and affordable. It is the AWS exam furthest from the word that sells courses, and closest to the work teams actually pay for.
A hundred and thirty minutes, 65 questions, offered in fewer languages than the older associate exams.
Exam facts
- Exam code
- DEA-C01
- Level
- Associate
- Field
- Data & AI
- Duration
- 130 minutes
- Questions
- 65 questions (50 scored, 15 unscored), multiple choice and multiple response
- Passing score
- 720 of 1,000 (scaled)
- Languages
- English, Japanese, Korean, Chinese
- Price
- 150 USD (read 12 September 2026)
- Validity
- 3 years, renewable
- Exam delivery
- Pearson VUE (test centre or online proctored)
Prerequisites: None. AWS recommends two to three years of data-engineering experience and one to two years of hands-on work with AWS data services.
Renewal: Pass the current version again (50 % voucher in your AWS certification account) or pass a higher-tier AWS exam
Source: exam page at AWS · Price: AWS quotes exam prices in US dollars; Pearson VUE charges the local equivalent at booking using the rate AWS sets
How to prepare
1. The vendor's free learning path
AWS publishes the exam objectives and a learning path free of charge — the authoritative source for scope and weighting. Open the learning path (opens in a new tab)
2. Online courses
Courses and practice questions from the directory that target exactly this exam — details, price and the provider link (affiliate) are on the course page.
Book the exam
Delivered by Pearson VUE (test centre or online proctored). Schedule with AWS (opens in a new tab)
Affiliate disclosure: the course links above is affiliate links — buying through them may earn us a commission at no extra cost to you. The vendor's learning path and booking links carry no commission.
The exam takes data quality seriously, which is unusual
Most syllabuses in this area move quickly to the interesting part. This one lingers on cataloguing, schema evolution, partitioning, permissions and lineage, which is a fair reflection of where the days go. Questions frequently describe a pipeline that already works and ask what should change when volume grows tenfold, when a schema shifts underneath it, or when an auditor asks who can read a column. If that sounds like your week, the exam will read as familiar.
- Choosing between batch and streaming for a described latency requirement, where the cheaper answer is usually correct.
- Partitioning and file format decisions whose cost effect is invisible until somebody queries a year of data.
- Permissions on data rather than on infrastructure, which is a distinct model and the part infrastructure people get wrong.
- Orchestration and failure: what happens to the run that half-finished, and how the next one knows.
- Catalogue and lineage, the questions an audit produces and nobody prepares for.
Four domains, and the first is a third of the paper
| Domain | Share | The recurring question |
|---|---|---|
| Data ingestion and transformation | 34 % | How does it get in, and what happens to it on the way |
| Data store management | 26 % | Where does it live, in what shape, and what does that cost to query |
| Data operations and support | 22 % | What broke overnight, and how does the next run know |
| Data security and governance | 18 % | Who may read which column, and how do you prove it |
Who this is written for
AWS recommends two to three years of data-engineering experience and one to two years of hands-on work with its data services, and the recommendation describes the paper accurately. The 65 questions, of which fifteen carry no score, run 130 minutes and need 720 of 1,000, and almost every one describes a working pipeline rather than a blank page. Analysts moving upstream will recognise the shapes and be vague on the operational half; engineers from another stack will know the concepts and reach for the wrong service name. Both should spend their preparation on the storage and governance domains, which are where a background on one side of the pipeline leaves the other side dark.
No practice test exists for this exam
Our practice-test partner had no product for this paper when we checked, twice, and rather than link an approximation we say so. The vendor's exam guide and free digital training cover the syllabus; a mock sitting in the exam's format does not exist in our catalogue for this exam. It is also offered in four languages only: English, Japanese, Korean and Simplified Chinese, which is a real constraint for European candidates.
Like every AWS credential this one lasts three years; renewal means resitting the current version at half price with the voucher AWS issues, or passing a higher-tier paper, and the higher pass renews this one at no charge. There is no higher data exam at the professional tier, so in practice renewal for a data engineer means a resit or a sideways climb into the architect or DevOps professional papers. Decide which before the first sitting, because it changes what you study over the three years in between.
What a question looks like
Written by us in the exam's style. It is not a real question from any question bank, and we do not publish those.
A daily job writes several thousand small files into object storage. Analysts report that queries over the last year have become slow and expensive, though the volume of data is modest. What change addresses the cause?
The symptom points at query tuning and the cause is upstream, in how the data was written. Candidates who work on the analysis side reach for an engine setting; candidates who have operated a pipeline recognise the small-file problem and fix it where it is created. The exam asks this shape of question repeatedly.
What it costs to get and to keep
| Item | Amount | Note |
|---|---|---|
| Exam fee | 150 USD | US dollar list price, converted at booking by Pearson VUE (read 12 September 2026) |
| Retake | 150 USD | the full fee again after the waiting period (read 12 September 2026) |
| Practice test | not published | we found no product for this exam with our practice-test partner when we checked, so the vendor's own sample questions are the realistic option (read 12 September 2026) |
| Recertification every 3 years | not published | by examination only; passing a higher exam renews it instead (read 12 September 2026) |
How much preparation, from where you are
- You already build pipelines, on any platform
- The hardest part is done. What remains is mapping practices you have to the AWS service that implements them, and learning the governance tooling, which tends to be the gap for people who came from an engineering rather than an analytics background.
- You are an analyst who wants to move upstream
- A genuinely good destination, and a real climb. You will know the query side and very little about orchestration, permissions or cost at volume, which is most of the exam.
- You hold an architect certificate already
- Expect less overlap than the shared tier suggests. Architecture answers where things run; this exam answers how data moves and who may read it.
What passing this does not prove
- Machine learning of any kind. This is the pipeline that feeds a model, not the model.
- SQL beyond a working level, and no statistics at all.
- Whether the data you moved was correct. Correctness is the job; the exam checks the machinery.
- Anything portable to another cloud beyond the concepts.
Against the alternatives
- DP-700 — Microsoft Certified: Fabric Data Engineer Associate
- Microsoft's closest equivalent, built around a single integrated product rather than a catalogue of composable services. The choice follows your employer's stack, and the ideas transfer better than the tooling does.
- SAA-C03 — AWS Certified Solutions Architect – Associate
- Broader, better known to recruiters, and less specific about what you can do. If a filter is your obstacle take that one; if the work is your obstacle take this.
- AIF-C01 — AWS Certified AI Practitioner
- A tier below and a different subject. Sitting that one first is optional and, for anyone doing this work, largely decorative.
DEA-C01 — quick answers
Is my language supported?
Fewer than for the older associate exams: English, Japanese, Korean and Simplified Chinese. That is a real constraint for European candidates, and the additional-time request for non-native English speakers has to be arranged before booking.
Do I need to know Spark?
Familiarity helps because the managed services wrap it, but the exam asks about the AWS surface rather than the engine underneath. Understanding what a distributed transformation costs and where it fails matters more than API detail.
How does this sit against a data-science path?
They are adjacent and not interchangeable. This certifies that data arrives reliably, is catalogued and is governed. Modelling starts after all of that, and teams generally hire for the two separately.
Why is governance a fifth of the paper?
Because it is a fifth of the job, and the fifth that gets an organisation fined. Column-level permissions, lineage and the catalogue are the questions an auditor asks and the questions a pipeline built in a hurry cannot answer. The exam treats them as engineering rather than paperwork, which is correct.
Is there a higher AWS data exam to climb to?
Not at the time of writing. The professional tier holds the architect and DevOps papers, either of which renews this certificate, so a data engineer who wants to keep it current without resitting has to climb sideways. That is a fair reason to hold this one for one cycle and decide afterwards.
What comes next
Concepts this exam draws on
Glossary entries with the reason each one matters for DEA-C01.
- Data Engineering
The whole exam is the data-engineering lifecycle on AWS: ingest, transform, store, operate.
- Data Wrangling
Glue, Lambda and Spark transformations are the wrangling questions in AWS clothing.
- Object Storage
S3 as the lake: layouts, formats, lifecycle rules and access patterns.
- SQL
Transformation and querying in the AWS data services are SQL underneath, and the exam shows you statements.
Last reviewed 12 September 2026 · Getting Digital