Study Guide

Databricks Certified Generative AI Engineer Associate Study Guide

Use the saved domain outline to connect design applications, data preparation, application development, assembling and deploying applications to scenario-based questions and explanations.

Download App Free Practice Exam Key Terms Glossary

How the Exam Is Structured

Databricks Certified Generative AI Engineer Associate (GenAI Associate) validates design applications, data preparation, application development, assembling and deploying applications. The ExamPal practice bank includes 322 premium questions and 40 free questions mapped across the official blueprint.

Domain	Weight	Focus
Domain 1: Design Applications	14%	Design a prompt that elicits a specifically formatted response; Design prompts for specific response formats
Domain 2: Data Preparation	14%	Apply a chunking strategy for a given document structure and model constraints; Filter extraneous content in source documents that degrades quality of a RAG application
Domain 3: Application Development	30%	Select Langchain/similar tools for use in a Generative AI application; Qualitatively assess responses to identify common issues such as quality and safety
Domain 4: Assembling and Deploying Applications	22%	Code a chain using a pyfunc model with pre- and post-processing; Control access to resources from model serving endpoints
Domain 5: Governance	8%	Use masking techniques as guard rails to meet a performance objective; Use masking techniques as guard rails
Domain 6: Evaluation and Monitoring	12%	Select an LLM choice (size and architecture) based on a set of quantitative evaluation metrics; Select the best LLM based on the attributes of the application to be developed

14% of exam

Domain 1: Design Applications

This section covers how to design LLM-enabled applications in Databricks by translating business requirements into prompts, model tasks, chain components, and AI pipeline inputs/outputs. It also includes selecting and ordering tools for multi-stage reasoning and deciding when to use Agent Bricks capabilities.

Design a prompt that elicits a specifically formatted response

Design prompts for specific response formats

Select model tasks to accomplish a given business requirement

Match model tasks to business needs

Select chain components for a desired model input and output

Choose chain components for inputs and outputs

Translate business use case goals into a description of the desired inputs and outputs for the AI pipeline

14% of exam

Domain 2: Data Preparation

This section covers preparing source data for retrieval-augmented generation (RAG) workflows, including chunking, filtering noisy content, selecting extraction tools, and loading chunked text into Delta Lake tables in Unity Catalog. It also addresses source document selection, retrieval evaluation, advanced chunking strategies, and the role of re-ranking in retrieval systems.

Apply a chunking strategy for a given document structure and model constraints

Filter extraneous content in source documents that degrades quality of a RAG application

Choose the appropriate Python package to extract document content from provided source data and format

Define operations and sequence to write given chunked text into Delta Lake tables in Unity Catalog

Identify needed source documents that provide necessary knowledge and quality for a given RAG application

Use tools and metrics to evaluate retrieval performance

Design retrieval systems using advanced chunking strategies

30% of exam

Domain 3: Application Development

Covers practical skills for building generative AI applications, including tool selection, prompt construction, retrieval design, model selection, guardrails, and evaluation/monitoring. It also includes agentic and multi-agent system development using MLflow, Agent Framework, Genie Spaces, and conversational APIs.

Select Langchain/similar tools for use in a Generative AI application

Qualitatively assess responses to identify common issues such as quality and safety

Select chunking strategy based on model & retrieval evaluation

Augment a prompt with additional context from a user's input based on key fields, terms, and intents

Create a prompt that adjusts an LLM's response from a baseline to a desired output

Implement LLM guardrails to prevent negative outcomes

Select the best LLM based on the attributes of the application to be developed

22% of exam

Domain 4: Assembling and Deploying Applications

Covers how to assemble AI applications using chains, retrieval, vector search, and model serving patterns. It also includes deployment, CI/CD, prompt lifecycle management, MCP server integration, and user-facing interfaces for agent scenarios.

Code a chain using a pyfunc model with pre- and post-processing

Control access to resources from model serving endpoints

Code a simple chain according to requirements

Choose the basic elements needed to create a RAG application: model flavor, embedding model, retriever, dependencies, input examples, model signature

Create and query a Vector Search index

Identify how to serve an LLM application that leverages Foundation Model APIs

8% of exam

Domain 5: Governance

This section covers governance practices for GenAI applications, with emphasis on guardrails, masking techniques, and mitigation strategies that support performance objectives and reduce risk. It also addresses protecting applications from malicious user inputs and ensuring data sources comply with legal and licensing requirements.

Use masking techniques as guard rails to meet a performance objective

Use masking techniques as guard rails

Select guardrail techniques to protect against malicious user inputs to a Gen AI application

Select guardrail techniques

Use legal/licensing requirements for data sources to avoid legal risk

Use legal/licensing requirements

Recommend an alternative for problematic text mitigation in a data source feeding a GenAI application

12% of exam

Domain 6: Evaluation and Monitoring

Covers how to evaluate LLMs and agents, choose metrics, and monitor deployed applications in Databricks. It also includes cost control, inference logging, AI Gateway, custom scorers, and incorporating SME feedback to improve performance.

Select an LLM choice (size and architecture) based on a set of quantitative evaluation metrics

Select the best LLM based on the attributes of the application to be developed

Select an embedding model context length based on source documents, expected queries, and optimization strategy

Configure vector search for a particular solution based on number of embeddings, update frequency, latency, and cost requirements

Select key metrics to monitor for a specific LLM deployment scenario

Evaluate agent performance using MLflow scoring and tracing

Use inference logging to assess deployed RAG application performance

Key Terms to Know

These terms are loaded from the shared terminology pack and appear across the question explanations.

AI Gateway: A Databricks capability that includes Inference Tables, Usage Tables, and rate limiting for tracking deployed LLMs or agents.
Agent Bricks: A Databricks feature set for solving problems with specialized agent patterns, including Knowledge Assistant, Multiagent Supervisor, and Information Extraction.
Agent Framework: A Databricks framework used to deploy and track LLMs or agents, including with AI Gateway.
CI/CD: Continuous integration and continuous delivery/deployment practices used here for updating indexes, promoting prompts, and testing components.
Databricks App: An application built on Databricks that can provide an interactive user-facing interface. In the question, it is used for customer support agents asking questions and receiving grounded answers.
Databricks Certified Generative AI Engineer Associate: A Databricks certification exam that assesses the ability to design and implement LLM-enabled solutions using Databricks, including RAG applications, LLM chains, model selection, governance, deployment, and monitoring.
Databricks Secrets: A Databricks feature for securely storing sensitive values such as API keys. In the question, it is used to store the external MCP server API key.
Delta Lake: A Databricks storage layer used here as the destination for writing chunked text tables in Unity Catalog.
Foundation Model APIs: Databricks APIs used to serve LLM applications leveraging foundation models.
Hugging Face Transformers: A related online tool/service used for working with transformer-based models in generative AI applications.
Inference Tables: Tables used to track inference activity for deployed models or agents.
Information Extraction: An Agent Bricks option used to extract structured information from content.
Knowledge Assistant: An Agent Bricks option used to solve problems by providing knowledge-based assistance.
LLM: A large language model used in generative AI applications; the exam expects knowledge of current LLMs, their capabilities, and how to select them for tasks.
LLM chains: Multi-step application flows that combine an LLM with other components such as tools, retrievers, or prompts to produce an output.
LLM-as-a-judge: An evaluation approach where a language model scores or judges model outputs instead of, or in addition to, human raters. In the question, it is proposed as a way to rescore responses.
LangChain: A tool used in generative AI applications for building chains and related workflows.
MCP: Model Context Protocol. In the question, MCP servers are integrated to give an agent access to external and managed data sources.

Official Materials and Guidance

This page is built from Databricks official materials and ExamPal shared release pack, the shared syllabus, topic tree, terminology pack, free pack, and premium pack.

-Databricks Genai Associate Exam Guide
-Guidance: Official Databricks exam guide PDF with sample questions
-Domain outline: Design Applications 14%; Data Preparation 14%; Application Development 30%; Assembling/Deploying Applications 22%; Governance 8%; Evaluation/Monitoring 12%.

Download App Official source Start Free Practice Exam