The short version
The Databricks Certified Associate Developer for Apache Spark exam checks that you can do basic data work with the Spark DataFrame API in Python and that you understand how Spark runs your code. It has 45 scored multiple-choice questions and a 90-minute limit. This guide follows the exam guide that Databricks says covers the live version as of October 8, 2026.
What the exam is
Databricks describes the exam as a test of two things. The first is your grasp of Spark architecture and its components. The second is your ability to use the DataFrame API for everyday manipulation tasks inside a Spark session. Those tasks include selecting and renaming columns, filtering and sorting rows, aggregating, handling missing data, joining, reading and writing with schemas, and working with user-defined functions and Spark SQL functions.
On top of that, the exam asks about execution and deployment modes, lazy evaluation, shuffling, broadcasting, Structured Streaming, Spark Connect, and common tuning and troubleshooting techniques. Databricks says that a person who passes can be expected to complete basic Spark DataFrame tasks using Python.
Two details are easy to miss. All code snippets on the exam are in Python, so you do not need Scala. And no test aids are allowed, including the API documentation, so you have to remember method names, argument order, and the main configuration defaults.
The exam page does not print an exam code like “AZ-104”. We use the name of the exam and the slug `databricks-spark-associate` for it on this site.
Who it is for
The exam suits data engineers, analytics engineers, developers, and analysts who write PySpark or want to start. Databricks lists no prerequisites. It does highly recommend related training and about six months of hands-on experience with Apache Spark. If you have less than that, you can still prepare, but plan for more study hours and more time in a notebook. The prerequisites page covers what to know first.
Format at a glance
| Item | What Databricks says |
|---|---|
| Exam | Databricks Certified Associate Developer for Apache Spark |
| Scored questions | 45, multiple choice |
| Time limit | 90 minutes |
| Passing score | Not published |
| Delivery | Online proctored or at a proctored test center |
| Language of code | Python |
| Test aids | None, including API documentation |
| Validity | 2 years; recertify by taking the full live exam |
| Prerequisites | None required; training and six months of Spark experience highly recommended |
The exam format page covers scoring, retakes, and rescheduling.
The seven domains and their weightings
| Domain | Weighting |
|---|---|
| Apache Spark Architecture and Components | 20% |
| Using Spark SQL | 20% |
| Developing Apache Spark DataFrame/DataSet API Applications | 30% |
| Troubleshooting and Tuning Apache Spark DataFrame API Applications | 10% |
| Structured Streaming | 10% |
| Using Spark Connect to deploy applications | 5% |
| Using Pandas API on Spark | 5% |
Architecture and components
The pieces of a Spark application: driver, executors, cluster manager, cores and memory, the execution hierarchy of jobs, stages and tasks, partitions and shuffles, transformations versus actions, lazy evaluation, caching and storage levels.
Spark SQL
Reading and writing data sources such as files and JDBC, running SQL directly on files, save modes, saving to persistent tables, and registering DataFrames as temporary views.
DataFrame API
The biggest domain. Column and row manipulation, deduplication, aggregations, date handling, joins and unions, schemas for input and output, user-defined functions, broadcast variables, and accumulators.
Troubleshooting and tuning
Partitioning, repartitioning and coalescing, data skew, reducing shuffles, Adaptive Query Execution, and reading driver and executor logs to find out-of-memory problems.
The three small domains
Structured Streaming covers the micro-batch model, output modes and sinks, windows, and deduplication with or without a watermark. Spark Connect covers its features and the client, cluster, and local deployment modes. The pandas API on Spark covers its advantages and pandas UDFs.
Official source
For the vendor’s own description, see the Databricks exam page. It is the final authority if anything here differs.
More guides for DATABRICKS-SPARK-ASSOCIATE
- Spark Associate Study Plan: 4 and 8 Weeks
- Spark Associate Sample Questions
- Is the Databricks Spark Associate Hard?
- Spark Associate Renewal and Retakes
- Is the Databricks Spark Associate Worth It?
- Spark Associate Cheat Sheet: Key Facts
- DE Pro vs Spark Associate: Which First?
Next step
Read the domains explained page to see where your time should go. When you want to check yourself, our Databricks Spark Associate practice tests are five original practice sets of 45 questions, tagged by the seven exam domains.
Frequently asked questions
Do I need to know Scala?
No. Databricks says all code snippets on this exam are in Python.
Does the certification expire?
Databricks says it is valid for 2 years. To recertify you take the full exam that is live at that time.
Can I look things up during the exam?
No. Databricks says no test aids are provided, including API documentation.