#dea-c01

Thinking about the AWS Data Engineer Associate? Cool. Here are 9 things I wish someone had screamed at me before I registered.

1. It's NOT a SQL exam.

Everyone assumes "data engineer" means "SQL expert." The DEA-C01 barely tests SQL. It tests your knowledge of AWS data services — Glue, Athena, Redshift, Kinesis, EMR, Lake Formation, Step Functions. If you walk in thinking your SQL skills will carry you, you'll walk out with a failing score.

2. AWS Glue is the star of the show.

Glue appears in roughly 25-30% of questions either directly or indirectly. ETL jobs, Glue crawlers, Glue Data Catalog, Glue Studio — know them all. I didn't study Glue enough and it cost me on at least 8 questions.

3. The exam blueprint lies about domain weights.

The official breakdown says:

Data Ingestion and Transformation (34%)

Data Store Management (26%)

Data Operations and Support (22%)

Data Security and Governance (18%)

In practice, Data Ingestion felt more like 40%. Security questions were harder than expected. The weights are approximate at best.

4. Kinesis will make you cry.

Kinesis Data Streams vs Kinesis Data Firehose vs Kinesis Data Analytics. They sound similar. They do different things. AWS loves testing whether you know which Kinesis service to use for which scenario. Here's the cheat sheet:

Data Streams: Real-time, you manage consumers, custom processing

Firehose: Near real-time, serverless delivery to S3/Redshift/etc

Data Analytics: SQL queries on streaming data (now part of Managed Apache Flink)

5. You need to understand data lakes deeply.

S3 as a data lake, Lake Formation for governance, Glue Data Catalog for metadata, Athena for querying. This isn't one question — it's an entire pattern that appears across dozens of questions. If you can't draw the data lake architecture from memory, keep studying.

6. The passing score is 720, not 700.

Unlike most AWS certs that use 720, some people still assume 700. The DEA-C01 requires 720/1000. That extra 20 points matters when you're on the edge.

7. Redshift isn't just "a data warehouse."

You need to know Redshift Spectrum (query S3 data), Redshift Serverless (no cluster management), distribution styles (KEY, EVEN, ALL), sort keys, and concurrency scaling. One question about distribution styles can be worth 2-3 points if you know it cold.

8. Practice exams are non-negotiable.

I'm not saying this because I'm trying to sell something. I'm saying this because the question format is unlike any other AWS exam I've taken. They give you complex scenarios — "Your company has 50TB of daily log data flowing from Kinesis into S3, and analysts need query results within 5 seconds" — and you need to pick the right architecture.

Free DEA-C01 practice questions were what finally got my scores from 65% to consistently above 80%. The $4.99 lifetime access on ExamCert is genuinely the best deal I've found. Money-back guarantee if you fail.

9. The cert is worth more than you think.

Data engineering is one of the fastest-growing specialties in tech. Glassdoor puts AWS-certified data engineers at $130K-$165K average. The demand is ridiculous because every company has data problems and not enough people who know how to solve them with cloud services.

The DEA-C01 proves you're not just someone who writes SQL queries — you're someone who designs end-to-end data pipelines in AWS. That distinction matters at the hiring table.

My timeline: 6 weeks of focused study, about 12 hours per week. Would recommend having at least AWS Cloud Practitioner or Solutions Architect Associate first. Going in completely cold would be painful.