The short version
The Databricks Certified Data Engineer Professional exam checks whether you can build, tune, secure, and ship production data pipelines on the Databricks platform. It is multiple choice, 120 minutes, and Databricks does not publish a passing score. This guide follows the exam guide that Databricks dated October 9, 2026, which is the version that replaced the earlier outline.
What the exam is
Databricks describes the exam as an assessment of advanced skills in building, optimizing, and maintaining production-grade data engineering solutions. The word that matters is production. Plenty of people can write a notebook that transforms a table. The exam asks what happens when that work has to run every night, stay cheap, stay governed, recover from failures, and deploy the same way to dev, staging, and prod.
Expect questions that put two or three reasonable options side by side and give you a constraint, such as latency, cost, or a privacy rule, that decides between them. Knowing the vocabulary is the entry fee. Choosing the right tool for a given situation is the actual test.
Who it is for
Databricks names data engineers who work with Unity Catalog, Auto Loader, Lakeflow Pipelines, Lakeflow Jobs, Lakehouse Federation, sharing, and the medallion architecture. The guide recommends one year of hands-on experience with the tasks in its outline. Nothing is required before you book, but Databricks highly recommends related training.
If you have only used Databricks for ad hoc analysis, this is a big step. If you already run pipelines and jobs there, most of the exam will feel like work you have seen, plus a handful of newer features worth a weekend of practice.
Format at a glance
| Item | What Databricks says |
|---|---|
| Level | Professional |
| Question type | Multiple choice; no test aids allowed |
| Time | 120 minutes |
| Scored questions | 60 in the October 9, 2026 exam guide; the certification page printed 59 when we checked |
| Unscored questions | May appear, are not marked, and extra time is already included |
| Delivery | Online proctored or at a test center |
| Languages | English in the guide; the certification page also lists Japanese, Portuguese (BR), and Korean |
| Passing score | Not published |
| Validity | 2 years, then you take the current exam again |
The exam format page explains each row in more detail, including retakes and rescheduling.
The nine sections and their weightings
The guide splits the exam into nine sections. The weightings add up to 100%.
| Section | Weighting |
|---|---|
| 1. Developing Code for Data Processing using Python and SQL | 23% |
| 2. Data Ingestion & Acquisition | 12% |
| 3. Data Manipulation | 12% |
| 4. Monitoring and Alerting | 10% |
| 5. Cost & Performance Optimization | 15% |
| 6. Ensuring Data Security and Compliance | 8% |
| 7. Data Governance | 5% |
| 8. Debugging and Deploying | 10% |
| 9. Data Modeling | 5% |
Section 1 is nearly a quarter of the exam. It covers Declarative Automation Bundles (formerly Databricks Asset Bundles), user-defined functions, streaming pipelines, Lakeflow Jobs, CDC with AUTO CDC, and Structured Streaming state. Add Cost & Performance and the three biggest sections hold just over half of the exam. Governance and Data Modeling are small, but they are also the easiest places to pick up points with a day of focused study.
Our domains explained page walks through all nine with typical traps.
Check which outline your notes follow
The guide carries two dates, “Exam Published” and “Exam Guide Updated”, and both read October 9, 2026. Older study notes may follow an earlier outline with different section names and weightings. In the current guide, sharing and Clean Rooms sit under ingestion, and a Data Manipulation section covers window functions, the VARIANT type, AI functions, and data quality expectations. Before you trust any weighting you find online, compare it with the guide itself. Databricks also asks candidates to re-read the guide two weeks before the exam date, because it is updated whenever the exam changes.
Official source
The Databricks certification page and the exam guide are the authority for every fact above: Databricks Data Engineer Professional certification page.
More guides for DB-DE-PRO
- Databricks Data Engineer Pro Study Plan
- Databricks Data Engineer Pro Sample Questions
- Is the Databricks Data Engineer Pro Hard?
- Databricks Data Engineer Pro Prerequisites
- DE Pro Renewal and Retakes Explained
- Is Databricks Data Engineer Pro Worth It?
- DE Pro Cheat Sheet: Key Facts to Review
- DE Pro vs Spark Associate: Which First?
- DP-750 vs Databricks Data Engineer Pro
Next step
Read the exam format page so nothing about the day surprises you. When you want to test yourself, our five original practice sets have 60 questions each, follow the nine sections and weightings of this guide, and report your score by domain. Pair them with Databricks documentation and hands-on work.
FAQ
Is there a prerequisite exam?
No. Databricks requires nothing before you book, although it highly recommends related training and about a year of hands-on work.
How long is the certification valid?
Two years. To recertify, you take the current version of the exam again.
Do I need the Associate exam first?
Databricks does not require it. The Associate is a separate, lower-level exam in the same track, and many people use it as a stepping stone.
Can I use documentation during the exam?
No. The guide says no test aids are allowed.