What it is
A proctored, 45-question multiple-choice exam that Databricks uses to certify practitioners who can build and ship generative AI applications on its platform, including retrieval-augmented generation (RAG) systems, tool-using agents and evaluation pipelines. It sits at "Associate" level in Databricks' certification ladder.
Syllabus
The exam guide splits questions across six domains: Design Applications (14%), Data Preparation (14%), Application Development (30%), Assembling and Deploying Apps (22%), Governance (8%), and Evaluation and Monitoring (12%). Databricks-specific tools tested include Vector Search, Model Serving, MLflow and Unity Catalog, plus prompt engineering and chaining techniques. Machine-learning code in the exam is in Python; non-ML data tasks may use SQL.
Format and time commitment
90 minutes, 45 scored multiple-choice questions (unscored items may also appear), no reference material or test aids allowed. It's a one-off exam, not a taught course: candidates prepare independently or via Databricks' own training material.
Prerequisites and level
No formal prerequisite, but Databricks recommends "6+ months of hands-on experience performing the generative AI solutions tasks outlined in the exam guide" plus related training.
Price and booking
$200 USD, booked through the Webassessor platform, delivered online (remote proctoring) or at a physical test centre. The certification is valid for 2 years before recertification is required.
