The SnowPro Advanced: Data Engineer Certification validates advanced expertise in building, optimizing, and operating data pipelines on Snowflake. It focuses on topics that map directly to day-to-day data engineering work, including Streams, Tasks, Snowpipe, Dynamic Tables, and semi-structured data. This article covers the exam overview, scoring domains, key study points for each topic, and a strategy to pass.
| Item | Details |
|---|---|
| Exam code | DEA-C02 |
| Questions | 65 (multiple choice, multiple select, and interactive) |
| Duration | 115 minutes |
| Passing score | 750 / 1000 |
| Exam fee | $375 USD |
| Prerequisites | Valid SnowPro Core certification |
| Delivery | Pearson VUE (test center or online proctored) |
| Certification validity | 2 years |
| Recommended experience | 2+ years of hands-on data engineering on Snowflake |
These are the five domains on the English official exam page and Study Guide (DEA-C02, last updated March 6, 2026).
| Domain | Weight | Key topics |
|---|---|---|
| 1.0 Data Movement | 28% | Loading and ingesting data (file formats, INFER_SCHEMA, stages and storage integrations), troubleshooting ingestion errors, continuous pipelines (Snowpipe, Snowpipe Streaming, streams, tasks, dynamic tables), connectors (Kafka, Spark, Python), data sharing and the Marketplace, external, Iceberg, and hybrid tables |
| 2.0 Performance Optimization | 19% | Troubleshooting underperforming queries; warehouse configuration (scale up vs. scale out, multi-cluster, Snowpark-optimized); clustering, materialized views, search optimization, query acceleration, and caching; ACCOUNT_USAGE and resource monitors; monitoring continuous pipelines (task history, alerts, data quality) |
| 3.0 Storage and Data Protection | 14% | Time Travel (including its impact on streams) and Fail-safe, cross-region and cross-cloud replication, analyzing micro-partitions with system functions (clustering depth, clustering keys), creating development environments with Time Travel and cloning |
| 4.0 Data Governance | 14% | Object tagging and data classification, data lineage, data quality monitoring, column-level security (Dynamic Data Masking, external tokenization), row access, projection, and aggregation policies, Data Clean Rooms |
| 5.0 Data Transformation | 25% | UDFs (Snowpark, SQL, JavaScript, UDTFs, UDAFs) and external functions, stored procedures and transactions, processing semi-structured and unstructured data (directory tables), Snowflake Cortex features, data transformations with Snowpark, development workflows with Notebooks, Git integration, and dbt Projects |
As of October 2026, the Japanese exam page and Study Guide still show the earlier weights (Data Movement 26%, Performance Optimization 21%; the other three domains are the same).
Streams are Snowflake's Change Data Capture (CDC) feature, tracking INSERT/UPDATE/DELETE operations against a table. Combined with Tasks, you can build schedule-based data pipelines.
| Stream type | Tracks | Use case |
|---|---|---|
| Standard | INSERT / UPDATE / DELETE | General-purpose CDC |
| Append-only | INSERT only | Append-only tables (logs, events) |
| Insert-only (for External Tables) | New file additions | Detecting new files on an External Table |
Tasks execute SQL on a schedule (CRON expression or minute-based interval) or in response to a Stream containing data (SYSTEM$STREAM_HAS_DATA). You can define parent-child relationships as a task tree to build pipelines in a DAG (directed acyclic graph) structure.
Snowpipe is a serverless data ingestion feature that detects new files arriving in a stage and automatically runs COPY INTO.
| Comparison | Auto-ingest Snowpipe | Snowpipe Streaming |
|---|---|---|
| Trigger | File arrival event | SDK API call |
| Latency | 1 to several minutes | Seconds |
| Requires files? | Yes (files on a stage) | No (row-level inserts) |
| Billing model | Serverless credits | Serverless credits |
Dynamic Tables let you define data pipelines declaratively on Snowflake. You define the target table with a SQL query, and it automatically refreshes as the source data changes.
Snowflake stores semi-structured data such as JSON, XML, Avro, Parquet, and ORC in the VARIANT type, and you can query it with SQL.
| Technique | When to apply | Effect |
|---|---|---|
| Clustering key | Tables of 1 TB or larger with frequent filters on specific columns | Better pruning rate |
| Materialized view | Frequently executed aggregation queries | Reuse of pre-computed results |
| Search Optimization Service | Equality lookups on high-cardinality columns | Faster point lookups |
| Warehouse sizing | Large queries with significant scan volume | Increased parallel processing capacity |
The Data Engineer exam tests more practical knowledge than the Core exam, so hands-on study is essential.
SnowPro Advanced: Data Engineer
Question 1
If you want to track only INSERT operations on a source table and ignore UPDATE/DELETE changes, which Stream type should you use?
Correct answer: B
An Append-only Stream tracks INSERT operations only and does not record UPDATE or DELETE changes. It is ideal for append-only workloads like log or event tables. A Standard Stream tracks all DML operations. The Insert-only Stream is for External Tables only. There is no Stream type called Delta Stream.
Is the SnowPro Core certification required for the SnowPro Advanced Data Engineer exam?
Yes. An active SnowPro Core certification is a prerequisite for the exam. It does not have to be COF-C03: a certification earned on COF-C02 also qualifies while it is still active. If your Core certification has expired, you must pass the Core exam again first. While you can take the Advanced exam immediately after passing Core, the Advanced exam builds on Core knowledge and demands a deep understanding of Streams, Tasks, Snowpipe, Dynamic Tables, and more, so we recommend allowing an additional 4-8 weeks of study time.
How does the exam test when to use Dynamic Tables versus Streams + Tasks?
The exam uses scenario-based questions to test the judgment call: choose Dynamic Tables when you want to declaratively define the pipeline, and choose Streams + Tasks when you need fine-grained control over CDC. Dynamic Tables shine for their simplicity — you declare the target table with a SQL query and data refreshes happen automatically. Streams + Tasks, on the other hand, offer the flexibility to handle intermediate change-data processing and error handling in detail.
How much of the exam covers semi-structured data?
There is no separate semi-structured data domain. In the English official Study Guide (DEA-C02, last updated March 6, 2026), objective 1.2 in Domain 1.0, Data Movement (28%), covers ingesting structured, semi-structured, and unstructured data and schema detection with INFER_SCHEMA, and objective 5.4 in Domain 5.0, Data Transformation (25%), covers transforming semi-structured data to structured data and back. Make sure you know how to store JSON/XML in the VARIANT type, access data via dot or bracket notation, expand nested arrays with the FLATTEN function and LATERAL JOIN, and query Parquet/ORC/Avro files (SELECT $1 FROM @stage syntax).
Practice with certification-focused question sets
Try free questionsNicheeLab Editorial Team
NicheeLab editorial team focused on data engineering and cloud certification learning. Content is structured around practical study needs and official exam domains.
Snowflake Certifications: All 11 Exams Explained (2026)
Every SnowPro certification on Snowflake's official list — C...
Snowflake Exam Difficulty Ranking: All 11 Certs Compared (2026)
All 11 SnowPro exams compared by difficulty in a ranking and...
Snowflake Study Guide: Fastest Pass Route by Exam (2026)
How to pass SnowPro certifications efficiently — official ma...
SnowPro Core (COF-C03): Complete Exam Guide (2026)
Pass the SnowPro Core exam — five domains, scope, sample que...
SnowPro Associate Platform (SOL-C01): Current Status & What to Take Next (2026)
As of October 2026, SnowPro Associate: Platform (SOL-C01) is...