Databricks-GenAI-Assoc Practice Test 23 — 15 Questions

Question 1

You are preparing a large dataset for fine-tuning a model using Databricks Delta Live Tables (DLT). Which configuration is best for ensuring data quality and lineage in this pipeline?

Accepted Answer

Implement DLT Expectations to flag or drop invalid records.. Delta Live Tables (DLT) with Expectations allows you to define constraints on data quality. By tagging data that fails these constraints, engineers ensure only clean data reaches the final training table. This provides built-in lineage and automatic quality monitoring. In the context of GenAI, clean data is paramount, as noise in the training set leads to poor model performance and unpredictable behavior in downstream applications.

Answer

Use standard Spark SQL jobs without DLT.

Answer

Write all data to a temporary folder and clean it after training.

Answer

Rely on the LLM to automatically filter out low-quality data during training.

Question 2

Refer to the exhibit. The developer is attempting to log a custom model to the Unity Catalog. Which configuration is missing to ensure the model is registered correctly under the specified Unity Catalog location?

Accepted Answer

mlflow.set_registry_uri("databricks-uc"). To log a model to Unity Catalog, the registry must be specified as 'databricks-uc'. The default behavior for 'databricks' tracking URI is to log to the workspace model registry. By explicitly setting the registry to the Unity Catalog destination, the developer ensures the model is governed by the correct security and lineage frameworks, allowing for seamless downstream deployment and access control within the enterprise environment.

Answer

mlflow.set_registry_uri("databricks")

Answer

mlflow.set_tracking_uri("databricks-uc")

Answer

mlflow.set_experiment("uc_experiment")

Question 3

A data scientist is building a GenAI application that uses the OpenAI API through Databricks external model endpoints. The application must use credentials stored securely in Databricks and must not expose the API key in code or logs. Which Databricks feature should the data scientist use to store and reference the OpenAI API key?

Accepted Answer

Databricks secrets with a secret scope. Databricks secrets provide a secure, encrypted store for sensitive credentials such as API keys. They can be referenced in code and in external model endpoint configurations without revealing the secret value. This meets the requirement to keep the OpenAI API key secure and out of code and logs.

Answer

Unity Catalog volumes with restricted permissions

Answer

Hardcoding the API key in a notebook and then deleting the cell

Answer

Environment variables set in the cluster's Spark configuration

Question 4

An engineer has a Databricks Vector Search index that stores chunk embeddings for a technical manual. Users report that queries containing exact part numbers return irrelevant chunks because the embedding model blurs numeric tokens. The engineer wants retrieval to consider both dense vector similarity and exact keyword matching on the same Delta table without building a separate search system. Which Databricks Vector Search feature should the engineer enable?

Accepted Answer

Hybrid search on the index. Hybrid search augments dense vector similarity with keyword matching, which restores recall for exact tokens like part numbers while preserving semantic matching for natural-language queries. The other options affect index freshness, capacity, or result volume, but none of them introduce the exact-match signal that the scenario requires.

Answer

A larger embedding model dimension

Answer

A higher number of returned results per query

Answer

Delta table change data feed on the source table

Question 5

When designing a production RAG application, which technique is most effective for preventing the LLM from hallucinating based on outdated information?

Accepted Answer

Enforcing a streaming data pipeline to keep the vector index updated.. Implementing a robust data pipeline that refreshes the vector index in real-time or near-real-time ensures that the retrieved context is current. When coupled with source attribution, this allows the system to verify findings against the latest data. This approach is essential for maintaining accuracy, as stale information in the vector database directly leads to hallucinations that can damage user trust in the AI application.

Answer

Increasing the temperature parameter of the model to maximum.

Answer

Restricting the LLM to a specific list of keywords for its output.

Answer

Adding a long system prompt instructing the model not to lie.

Question 6

A generative AI team stores prompt/response logs in a Unity Catalog Delta table named `main.genai.interaction_logs`. Compliance requires that the raw `prompt_text` column be readable only by members of the `ai_compliance` group, while all other users querying the table must see the literal string `REDACTED`. Which Unity Catalog feature should be implemented directly on the table to satisfy this requirement without creating a separate view?

Accepted Answer

Column mask. Column masks are the Unity Catalog mechanism designed to transform individual column values based on the caller's group membership while leaving the table itself queryable by everyone. Binding a mask function to prompt_text that checks is_account_group_member('ai_compliance') gives compliance staff the raw text and returns REDACTED to all other users, meeting the requirement on the existing table.

Answer

Row filter

Answer

Dynamic view with a CASE expression

Answer

Attribute-based access control tag

Question 7

When designing a RAG application for a regulated industry, what is the most important Unity Catalog feature for ensuring data compliance during retrieval?

Accepted Answer

Row-level security and column masking.. Unity Catalog's attribute-based access control (ABAC) or row-level security allows for granular control over which users can access specific pieces of data. In regulated industries, it is critical that the RAG model only retrieves information that the querying user is explicitly authorized to view, preventing sensitive data leakage and ensuring auditability at the catalog level.

Answer

Catalog-level tagging for organization.

Answer

Personal Access Tokens (PATs) for service principals.

Answer

Automatic data lineage tracking.

Question 8

Which TWO factors should be considered when choosing an embedding model for a RAG application?

Accepted Answer

The model's native dimensionality and its effect on vector index storage size.. The choice of embedding model determines the quality of semantic retrieval and the computational requirements of the system. By aligning the model's language support, dimensionality, and performance with the application's needs, you ensure a balanced design. It is also vital to consider the model's licensing and whether it needs to be fine-tuned to capture domain-specific terminology that general models might misinterpret.

Answer

The model's ability to generate creative, hallucinatory content.

Answer

The maximum number of characters allowed in a single prompt.

Answer

The color profile of the documents being ingested.

Question 9

An engineer is designing a GenAI application where a Databricks workflow must call an external LLM provider that enforces a strict rate limit and occasionally returns transient errors. The engineer wants the application to degrade gracefully instead of failing the whole job. Which design choice best addresses this requirement?

Accepted Answer

Wrap the external call in retry logic with exponential backoff and a fallback response path. Resilience to external rate limits and transient failures comes from handling the call itself: bounded retries with exponential backoff smooth over temporary rejections, and a fallback response keeps the application functional when the provider stays unavailable. Infrastructure sizing, task-level retries, and response caching do not address the failure at its source.

Answer

Increase the driver node type on the job cluster

Answer

Cache all prompts and responses in a Delta table before calling the provider

Answer

Set the job to retry the entire task on failure

Question 10

A Databricks team is building a retrieval corpus from mixed-format documents stored in a Unity Catalog volume. They need a preparation pipeline that preserves document structure for later chunking and that records which source file each chunk came from so retrieval results can cite evidence. Which two design choices best meet these requirements? (Choose two.)

Accepted Answer

Carry the source file path as a metadata column on each chunk and store it in the Delta table alongside the chunk text and embedding.. Preserving document structure requires a parser that extracts layout elements rather than flattening everything to plain text, and ai_parse_document provides that for PDFs and images. Provenance requires carrying the source file path on each chunk so retrieval results can cite the originating document. Together these choices keep structure intact and make every chunk traceable.

Answer

Concatenate all documents into a single large text field before chunking to reduce the number of rows.

Answer

Convert every document to plain text and discard page and heading markers to normalize the corpus.

Answer

Generate embeddings for entire documents rather than chunks to avoid storing multiple rows per file.

Question 11

You are preparing a large corpus of customer support emails stored as Parquet files in a Unity Catalog volume. The emails must be cleaned by removing boilerplate signatures and disclaimers before tokenization for a fine-tuning dataset. Your team wants to enforce this transformation as a declarative, testable pipeline stage that fails fast if any cleaned record still contains a known boilerplate marker. Which Databricks capability should you use to implement this cleaning step with built-in data quality expectations?

Accepted Answer

Define the cleaning logic in a Delta Live Tables pipeline and attach an EXPECT constraint that asserts no boilerplate marker remains in the cleaned column.. Delta Live Tables expectations are the declarative, testable mechanism for enforcing data quality during preparation. Attaching an EXPECT constraint to the cleaned table ensures the pipeline fails fast if any boilerplate marker remains, while capturing metrics for auditing. This approach provides lineage, observability, and reproducibility, unlike imperative notebooks or asynchronous alerts.

Answer

Use Spark Structured Streaming with a foreachBatch function that calls a Python UDF to strip signatures and raises an exception if a marker remains.

Answer

Create a Databricks SQL query that applies regexp_replace to the email body and schedule it as a SQL warehouse alert when a marker is detected.

Answer

Write a notebook that reads the Parquet files, applies a pandas UDF to clean the text, and writes the result back to the volume, then manually review the output.

Question 12

Which Databricks component is recommended for orchestrating the end-to-end RAG pipeline, including data ingestion, transformation, and vector indexing?

Accepted Answer

Databricks Workflows.. Databricks Workflows allows for the scheduled and triggered execution of tasks, such as notebook runs or Delta Live Tables pipelines. This is the recommended way to orchestrate complex RAG pipelines where data must be cleaned, chunked, and indexed systematically. By using Workflows, engineers ensure that the entire pipeline is reproducible, monitorable, and resilient to failures, which is essential for maintaining a production-grade AI application.

Answer

Databricks SQL.

Answer

The Databricks File System (DBFS).

Answer

Unity Catalog.

Question 13

A company is scaling its RAG applications. Which THREE of the following are benefits of using Databricks Vector Search over managing a standalone vector database outside of the platform?

Accepted Answer

Automatic, continuous synchronization with Delta tables.. Databricks Vector Search offers benefits through tight integration, including automatic synchronization with Delta tables, which removes the need for custom ETL. Because it resides in the same security boundary, it inherits Unity Catalog's governance and access controls, ensuring data security. Finally, it uses the platform's managed infrastructure, eliminating the overhead of maintaining external database clusters, which allows developers to focus on application logic rather than infrastructure maintenance.

Answer

Ability to use SQL as the only way to perform embedding calculations.

Answer

Support for non-Databricks proprietary vector file formats.

Question 14

An organization needs to build a RAG application on Databricks that minimizes data egress and maximizes security by keeping all data within the workspace perimeter. Which architectural pattern best satisfies this requirement?

Accepted Answer

Deploy an embedding model on Mosaic AI Model Serving and utilize Databricks Vector Search.. Utilizing Mosaic AI Model Serving with private endpoints and leveraging Vector Search indexes ensures that both the embedding model and the retrieval process occur within the Databricks control plane. By avoiding external API calls to third-party providers, the organization maintains strict governance, data residency compliance, and lower latency for inference, which is critical for enterprise-grade generative AI applications handling sensitive corporate documents.

Answer

Call external LLM APIs from a standard Python notebook without VPC constraints.

Answer

Export data to an external vector database and use a cloud-hosted LLM.

Answer

Use a public LLM endpoint with a public bucket to store the vector index.

Question 15

A GenAI engineer is developing an agent on Databricks that must call a custom Python function to query an inventory database. The agent must decide when to invoke the function and must receive the result back into its reasoning loop. Which TWO actions are required to expose the function as a tool the agent can call? (Choose two.)

Accepted Answer

Register the function with the agent using the framework's tool decorator or tool registration API so it appears in the tool list passed to the model.. Exposing a Python function as an agent tool requires two things: a well-formed schema the model can understand, derived from the function's name, docstring, and type hints, and explicit registration so the function is included in the tool list the agent passes to the LLM. Together these let the model decide when to call the tool and let the agent execute it and return results into the reasoning loop. The remaining options neither register the function nor execute it.

Answer

Deploy the function as a separate Mosaic AI Model Serving endpoint and hard-code its URL in the system prompt.

Answer

Add the function's source code to the retrieval index so the model can retrieve it as context at inference time.

Answer

Convert the function's return value into an embedding and store it in Vector Search for similarity lookups.