Databricks periodically refreshes its certification exam scope to keep pace with product rebrands and new features. 2026 brought sweeping changes across several exams, including renamed concepts, rebalanced domain weights, and an entirely new exam. This article walks through the major 2026 updates and what they mean for your test prep.
The table below summarizes the major changes that have already shipped or are scheduled for 2026.
| Exam | Change | Effective Date |
|---|---|---|
| Data Engineer Associate（DEA） | Domain renames, Unity Catalog weight 20% → 30%, DLT renamed to Lakeflow | January 2026 |
| Data Engineer Professional（DEP） | New coverage of Serverless Compute and Lakeflow Connect | February 2026 |
| Machine Learning Associate（MLA） | Feature Store → Feature Engineering in UC; Mosaic AI integration | January 2026 |
| Spark Developer Associate | Increased weight for Spark Connect and pandas API on Spark | March 2026 |
| GenAI Engineer Associate | Brand-new exam (officially released in H2 2025) | October 2025 onward |
| Data Analyst Associate（DAA） | New coverage of Genie and AI/BI Dashboards | February 2026 |
| Machine Learning Professional（MLP） | New coverage of Mosaic AI Model Serving and Lakehouse Monitoring | March 2026 |
DEA is one of the exams most affected by the 2026 update. Here are the main changes.
The old "Data Processing with Apache Spark" domain has been renamed to "ELT with Spark SQL and Python." The focus has shifted to data transformation using Spark SQL and Python, and low-level RDD operations have been dropped from the exam.
The "Data Governance" domain has grown from roughly 20% to about 30% of the exam, and Unity Catalog coverage has expanded significantly. Expect strong focus on the 3-level namespace (catalog.schema.table), external locations, storage credentials, and fine-grained access control via dynamic views.
Delta Live Tables (DLT) has been rebranded as "Lakeflow Declarative Pipelines." The exam uses the new name, but the API and code (e.g. the @dlt.table decorator) are unchanged. Similarly, workflow jobs have been renamed "Lakeflow Jobs."
Here is a mapping of the key terminology changes in the 2026 update. The exam uses the new names, but you may still see the old names in some documentation.
| Old Name | New Name (2026+) | Notes |
|---|---|---|
| Delta Live Tables（DLT） | Lakeflow Declarative Pipelines | API still uses @dlt.table — no code changes |
| Databricks Workflows / Jobs | Lakeflow Jobs | Job scheduler rename |
| Feature Store | Feature Engineering in Unity Catalog | Consolidated under UC; uses the databricks-feature-engineering package |
| Z-ORDER BY | Liquid Clustering (recommended) | Z-ORDER is deprecated; use the CLUSTER BY clause |
| MLflow Model Registry (Workspace edition) | MLflow Model Registry in Unity Catalog | Migration to the UC edition is recommended |
| Databricks Model Serving | Mosaic AI Model Serving | Includes Foundation Model APIs |
| Databricks Vector Search | Mosaic AI Vector Search | Same functionality |
| DBFS（Databricks File System） | Unity Catalog Volumes (recommended) | DBFS is legacy; migration to Volumes is recommended |
GenAI Engineer Associate is the newest Databricks certification, officially released in October 2025. It tests practical knowledge of building generative AI applications.
Since 2023, Databricks has rapidly expanded its generative AI capabilities under the Mosaic AI brand — Foundation Model APIs, Vector Search, AI Gateway, and more. GenAI Engineer Associate was created to validate the skills needed to build RAG applications and LLMOps pipelines on top of these features.
Here is the recommended approach to timing your exam around the 2026 update.
Once an exam update goes live, the official Exam Guide on Databricks Academy is refreshed. Read the latest Exam Guide first so you can see the new domain weights and any added or removed topics. The NicheeLab question bank is updated in sync with each revision, so you can prep with the most current questions.
After an update goes live, the exam uses the new scope. If you study from outdated books or question banks, you will be caught off guard by new topics like Lakeflow, Liquid Clustering, and Mosaic AI. Before you start studying, confirm whether the update has already taken effect for your target exam.
If you plan to earn several Databricks certifications, the most efficient sequence is DEA → Spark Developer → MLA → GenAI Engineer. DEA grounds you in the Databricks platform, Spark Developer sharpens your coding skills, MLA layers on machine learning, and GenAI covers the latest generative AI stack.
DEA - 2026 Update
問題 1
Which statement about the 2026 Databricks exam update is correct?
正解: A
Delta Live Tables (DLT) has been rebranded at the product level as "Lakeflow Declarative Pipelines," but there are no API-level changes. The @dlt.table decorator, @dlt.view decorator, and Expectations (data quality constraints) all continue to work as-is. B is wrong: Z-ORDER BY is deprecated in favor of Liquid Clustering, but it has not been removed. C is wrong: Feature Store has already been integrated into UC as "Feature Engineering in Unity Catalog." D is wrong: GenAI Engineer Associate has no prerequisites.
Do the 2026 updates invalidate existing Databricks certifications?
No. Existing certifications remain valid until they expire (2 years from the date earned). You will need to recertify under the updated exam, but only after expiration — certifications earned before the update are never invalidated early. That said, recertification exams use the new exam scope, so plan your study around the updated content.
Is study material based on the old DLT (Delta Live Tables) name now obsolete?
No — the technical content is identical. DLT was rebranded to "Lakeflow Declarative Pipelines," but the APIs and features are unchanged. The @dlt.table decorator, Expectations (data quality constraints), and the concepts of live tables and streaming tables all still apply. The exam will use the new name, so just memorize the old-to-new mapping and you are fine.
Are there prerequisites for the GenAI Engineer Associate exam?
No prerequisites. Every Databricks certification is independent, so you can start with any exam. That said, GenAI Engineer Associate tests specialized topics like RAG, Vector Search, and LLMOps, so a working knowledge of the Databricks platform and Python is recommended. It is usually more efficient to earn DEA or MLA first to ground yourself in the platform before tackling GenAI Engineer.
Practice with Questions Aligned to the Latest Update
Prep for the real exam with questions that reflect the 2026 update
Try free questions →Related Databricks Exam Articles
All Databricks Certifications
Complete guide to all 7 exams — scope and difficulty
Databricks Exam Sample Questions
Question formats and answer walkthroughs for all 7 certifications
GenAI Engineer Associate Exam Prep
Exam scope, study plan, and tips for passing
Unity Catalog: Complete Guide
The 3-level namespace and access control
Practice with certification-focused question sets
無料で問題を解いてみるNicheeLab Editorial Team
NicheeLab editorial team focused on data engineering and cloud certification learning. Content is structured around practical study needs and official exam domains.
Databricks Certifications: All 7 Exams, Difficulty & Study Plan (2026)
Complete guide to all 7 Databricks certifications — Data Eng...
Databricks Exam Difficulty Ranking: All 7 Certs Compared (2026)
Every Databricks certification ranked by difficulty, with st...
Databricks Study Guide: Fastest Pass Route & Time Estimates (2026)
How to pass Databricks certifications efficiently. Official ...
Databricks Data Engineer Associate: Complete Guide (2026)
Domain-by-domain breakdown of the Databricks Certified Data ...
Databricks Data Engineer Professional: Complete Guide (2026)
Tactics for the Databricks Certified Data Engineer Professio...