Databricks has two data engineering certifications: Data Engineer Associate and Data Engineer Professional. They share a fee, a question type and a validity period, but the Professional exam is longer, expects more experience and has just moved to a new outline. Here is how they compare, taken from Databricks’ certification pages and exam guides, followed by our suggestion on which to take first.
The two exams side by side
| Detail | DE Associate | DE Professional |
|---|---|---|
| Scored questions | 45, multiple choice | 60, multiple choice |
| Time | 90 minutes | 120 minutes |
| Passing score | Not published by Databricks | Not published by Databricks |
| Registration fee | USD 200, plus applicable taxes | USD 200, plus applicable taxes |
| Delivery | Proctored, online or at a test centre | Proctored, online or at a test centre |
| Test aids | None allowed | None allowed |
| Prerequisites | None required | None required |
| Recommended experience | Six months of hands-on experience in Databricks | One year of hands-on experience with the tasks in the outline |
| Sections | Seven | Nine |
| Valid for | 2 years | 2 years |
| Current exam guide | Exam published 4 May 2026 (guide updated 6 August 2026) | Exam published 9 October 2026 |
Both exams may also include unscored questions. Databricks uses them to gather statistical information, they are not identified, they do not affect your score, and extra time is already built into the time limit.
Who each exam is for
Associate. Databricks says the exam assesses your ability to use the Databricks Data Intelligence Platform for foundational data engineering tasks. That covers the platform and its architecture, ingestion and loading, transformation and modelling (including ETL with PySpark), Lakeflow Jobs, CI/CD, troubleshooting and optimization, and governance and security.
Professional. Databricks says this exam assesses advanced skills in building, optimizing and maintaining production-grade data engineering solutions. Candidates should be able to design secure, reliable and cost-effective ETL pipelines in Python and SQL. They are also tested on streaming, orchestration, caching, debugging, CI/CD, the Databricks CLI and REST API, AI functions and Declarative Automation Bundles.
On code, the Associate certification page says code is shown in SQL when possible and otherwise in Python. The Professional page says its code examples will primarily be in Python and SQL.
Recommended experience
Neither exam has a required prerequisite, and Databricks highly recommends the related training for both. The exam guides then set different experience levels:
- Associate: course attendance and six months of hands-on experience in Databricks are highly recommended.
- Professional: one year of hands-on experience in the data engineering tasks in the exam outline is highly recommended.
The certification pages word this more loosely for both exams: hands-on experience performing the tasks outlined in the exam guide, with no length of time given.
Associate sections and weights
| Associate section | Weight |
|---|---|
| Databricks Intelligence Platform | 6% |
| Data Ingestion and Loading | 21% |
| Data Transformation and Modeling | 22% |
| Working with Lakeflow Jobs | 16% |
| Implementing CI/CD | 10% |
| Troubleshooting, Monitoring, and Optimization | 10% |
| Governance and Security | 15% |
Professional sections and weights
This is the outline in the Professional exam guide dated 9 October 2026:
| Professional section | Weight |
|---|---|
| Developing Code for Data Processing using Python and SQL | 23% |
| Data Ingestion & Acquisition | 12% |
| Data Manipulation | 12% |
| Monitoring and Alerting | 10% |
| Cost & Performance Optimization | 15% |
| Ensuring Data Security and Compliance | 8% |
| Data Governance | 5% |
| Debugging and Deploying | 10% |
| Data Modeling | 5% |
The new Professional exam guide
The Professional exam guide is dated 9 October 2026 and lists the exam as published on that date. It has nine sections and 60 scored questions. Its sample questions follow the new outline, and Databricks notes that if your exam was on or before 9 October 2026, a few of them may cover objectives that were not on your exam.
Two official sources differ. On 16 October 2026 the Professional certification page still showed the older outline: ten sections with different weights, and 59 scored questions. The exam guide is newer, so we follow it. Databricks asks candidates to check the exam guide two weeks before their exam date.
The Associate exam guide covers the exam version as of 4 May 2026. On 16 October 2026 its certification page showed the same seven sections and weights.
Where the two exams overlap
The two outlines share a lot of ground. Both cover Auto Loader, Lakeflow Jobs, Unity Catalog permissions and ABAC policies, Liquid Clustering and predictive optimization, the Spark UI, Declarative Automation Bundles and Databricks Git folders.
The Professional outline goes further. Its objectives include:
- Structured Streaming watermarks, output modes, foreachBatch and checkpoints
- AUTO CDC with SCD Type 1 and Type 2, and CDC from SQL Server, MySQL and PostgreSQL with Lakeflow Connect
- OpenSharing, Clean Rooms and Lakehouse Federation
- the VARIANT type and AI functions such as ai_query
- system tables, pipeline event logs and alerts for monitoring
- PII detection and masking, and data purging for rules such as GDPR erasure
- unit and integration tests, and repairing failed job runs
- dimensional models with materialized views and Unity Catalog metric views
Both guides use current product names. Git folders were formerly Repos, and Declarative Automation Bundles were formerly Databricks Asset Bundles. The Professional guide also uses AUTO CDC, formerly APPLY CHANGES. Older study material may use the former names.
Pass marks
Databricks does not publish a passing score for either exam. Neither the certification pages nor the exam guides state one, so we do not give a figure here.
Keeping the certification
Both certifications are valid for two years. To recertify the Professional, you take the current live exam. For the Associate, the exam guide gives two options: the Data Engineer Associate Renewal Exam or the current full exam. Databricks updates its exam guides whenever an exam changes, so read the latest guide before you sit an exam again.
Our suggestion: which to take first
This is our suggestion, not Databricks’. Databricks does not require the Associate before the Professional.
- You have less than a year of hands-on Databricks work: start with the Associate. Its guide recommends six months of experience, against one year for the Professional, and its topics are a base the Professional builds on.
- You want both: take the Associate first, then the Professional. Much of what you learn for ingestion, jobs, CI/CD and governance carries over.
- You already run production pipelines on Databricks: going straight to the Professional can make sense. Check the outline first for gaps in streaming, CDC, sharing, monitoring with system tables and data privacy.
- You are preparing for the Professional now: use the 9 October 2026 exam guide, not the older outline, and plan for 60 questions in 120 minutes.
Confirm the latest details on Databricks’ Data Engineer Associate page and Data Engineer Professional page before you book.
I run ExamCrest, where this post first appeared. Try free practice questions for the Associate and the Professional.
