Exam prep for

Databricks Certified Generative AI Engineer Associate

Exam set and awarded by Databricks On demand

For practitioners preparing for the Databricks Certified Generative AI Engineer Associate credential: build the skills to design, deploy, govern and evaluate generative AI applications on Databricks.

Included with Elandi Pro: practice on all 6 exam domains and unlimited mock exams.

Independent exam prep. Elandi is not affiliated with, endorsed by or approved by Databricks, and does not award Databricks credentials. Our practice questions and mocks are written by Elandi, not taken from the real exam. Exam names are trademarks of their owners.

The exam in numbers

The exam assesses the design and implementation of LLM-enabled solutions using Databricks, including retrieval-augmented generation, deployment and governance. It uses multiple-choice and multiple-selection questions. For the Elandi mock, 70% is our practice target, not a Databricks pass mark. The 75% single-select and 25% multiple-select mix is our practice mix, not a published Databricks distribution. Four options for single-select and five for multiple-select follow the guide’s examples; neither is a confirmed universal exam rule. Estimated preparation time: 16–40 hours for an experienced practitioner (no official figure). The current Databricks catalog (https://www.databricks.com/training/catalog/generative-ai-engineering-with-databricks-1980) lists 16 hours for its recommended course; the upper end allows up to 24 additional hours for practice and review and is an Elandi planning assumption, not a measured historical average. Your mock exams use this learning plan's question count, time limit and pass mark.

Questions
45
Time limit
90min
Mock pass mark
70%
32 correct
Exam fee (paid to Databricks)
US$200
When you book the exam
Calculator
Not allowed
Sitting
On demand
  • Single select (Radio): 4 options
  • Multiple select (Checkbox): 5 options

Upcoming sittings

Pick a sitting when you start. Your workspace counts down to it and paces your prep from there.

  1. On demand Book any day

What Databricks Certified Generative AI Engineer Associate covers

6 modules · 29 units · ~29h of structured prep. Each exam domain with the weight the exam gives it, and the modules that prepare you for it.

  1. 01Design Applications 14%
    1. Prompts and model tasks
    2. Pipeline components and inputs
    3. Tools and Agent Bricks
  2. 02Data Preparation 14%
    1. Chunking and document cleanup
    2. Extraction and Delta ingestion
    3. Source selection and retrieval measurement
    4. Advanced chunking and re-ranking
  3. 03Application Development 30%
    1. Framework tools and response quality
    2. Chunking from evaluation results
    3. Context and prompt improvement
    4. Guardrails and LLM selection
    5. Embedding context and model metadata
    6. Experiment metrics and agent development
    7. Lifecycle phases and data-aware agents
  4. 04Assembling and Deploying Applications 22%
    1. Pyfunc chains and endpoint access
    2. Chains and RAG components
    3. Model registration and index operations
    4. Foundation model serving and AI Search
    5. Batch inference and search configuration
    6. Persistent memory and delivery workflows
    7. MCP integration and prompt lifecycle
    8. Interactive agent interfaces
  5. 05Governance 8%
    1. Masking and input guardrails
    2. Source licensing and problematic text
  6. 06Evaluation and Monitoring 12%
    1. Model selection and monitoring metrics
    2. Agent scoring and inference logging
    3. Cost controls and live endpoint monitoring
    4. Judges and AI Gateway monitoring
    5. Custom scorers and expert feedback

Your exam workspace counts down and keeps score

Set your exam day once. The workspace paces your prep in four phases, weighs your coverage by what each topic is worth, and shows where the marks are still open after every mock.

Exam in 42 days Start mock exam
  • Overview
  • All attempts
  • Review
  • Preview questions
Questions
45
Time limit
90 minutes
Pass score
70% · 32 correct
42

days until your exam

6 weeks

Saturday, 2 January 2027

  1. Learn 8+ weeks out
  2. Mock and repair 8 to 2 weeks out
  3. Simulate Last 2 weeks
  4. Consolidate Last 3 days

Now: Mock and repair. Sit a full mock every week or two, then practise the topics you missed.

How ready you are

Exam-weighted coverage
70% of the plan, counted by what each topic is worth
Full mocks taken
2 Prep guides suggest three to five before the exam
Latest mock
66% 4 points below the 70% pass mark

Where the marks are still open

  1. Assembling and Deploying Applications 22% of the exam 54%
  2. Application Development 30% of the exam 71%
  3. Evaluation and Monitoring 12% of the exam 47%
  4. Governance 8% of the exam 63%
  5. Design Applications 14% of the exam 88%
  6. Data Preparation 14% of the exam 100%

How the content is written

Practical, precise exam preparation for generative AI practitioners, focused on choosing and evaluating Databricks application designs. Explain decisions and trade-offs with concise examples rather than introductory platform tutorials.

Worked examples
Some
Technical jargon
Some
Maths & formulas
Light
Code
Light
Diagrams
Light

Try a sample question

Elandi wrote these in this learning plan's style. Pick an answer to see whether it's right and why. They are not taken from the exam.

Select one answer
  • A team is comparing two versions of an agent before release. It has a fixed evaluation dataset and agreed quality scorers, and needs a repeatable comparison of the versions on the same cases. Which approach should the engineer use? One answer
    Select one answer
  • A team is deploying a fine-tuned variant of a Databricks-hosted foundation model for an interactive application. The workload requires performance guarantees. Which serving approach best fits these requirements? One answer
    Select one answer
  • A RAG application uses an AI Search index built from a Delta table. The team adds a source column that must be returned in search results, while keeping the existing application available throughout the change. What should the engineer do? One answer
    Select one answer
  • A team maintains a Delta table for a search application. It wants an AI Search Delta Sync Index that processes only changed rows on each sync and is prepared to use a standard endpoint. Which TWO configurations satisfy these requirements? Select two. Select all that apply
    Select all that apply

Start your Databricks Certified Generative AI Engineer Associate learning plan.

Start preparing for Databricks Certified Generative AI Engineer Associatewith Elandi Pro today.