Generative AI Leader Practice Exam 60 — 20 Questions

Question 1

A media company is evaluating Google Cloud's generative AI offerings for two distinct needs: enabling its journalists to summarize research inside Google Docs, and building a custom internal tool that calls Gemini models programmatically with its own authentication and logging. Which two Google Cloud offerings map to these needs? (Choose two.)

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Vertex AI for programmatic Gemini model access. The journalists' in-editor summarization need is met by Gemini for Google Workspace, while the custom programmatic tool requires Vertex AI's APIs, IAM, and logging. The remaining services handle speech, conversational flows, or document parsing and do not provide the requested generative capabilities.

Answer

Document AI for parsing research documents

Answer

Cloud Speech-to-Text for transcribing interviews

Answer

Dialogflow CX for building the internal tool

Question 2

A small marketing agency wants to experiment with Google's generative AI models without writing code or managing cloud infrastructure. They need a browser-based environment to draft campaign ideas and test prompts quickly. Which Google Cloud offering should they use?

Accepted Answer

Vertex AI Studio. Vertex AI Studio is the console-based workspace for prompt design and model testing, so a non-technical marketing team can draft and compare campaign ideas in a browser without code. The other choices require SQL, pipeline definitions, or deployed serverless code, none of which match a quick, no-code ideation need.

Answer

BigQuery ML with remote model inference

Answer

Vertex AI Pipelines

Answer

Cloud Run functions calling the Gemini API

Question 3

A hospital network wants to build a generative AI search experience over its clinical guideline PDFs so clinicians can ask natural-language questions and receive answers with citations to the source documents. The solution must run on Google Cloud and keep data within the network's project. Which offering is purpose-built for this requirement?

Accepted Answer

Vertex AI Search with grounding on the guideline corpus. Vertex AI Search indexes the guideline documents and returns grounded answers with citations, which is exactly the retrieval-augmented experience described. Vision, Workspace, and Translation services perform extraction, productivity assistance, or language conversion and cannot deliver cited question answering over the hospital's corpus.

Answer

Gemini for Google Workspace

Answer

Cloud Vision API with OCR on the PDFs

Answer

Cloud Translation API

Question 4

A financial analytics firm wants to prototype prompts against several Gemini model versions quickly, compare outputs side by side, and then export the winning prompt configuration into a production application with enterprise controls. Which combination of Google Cloud offerings best supports this workflow?

Accepted Answer

Google AI Studio for prototyping, then Vertex AI for production deployment. Prototyping prompts against multiple Gemini models is fastest in Google AI Studio, and moving the validated configuration into a governed environment is what Vertex AI provides. The productivity suite and the language analysis API cannot perform side-by-side generative prompt comparison or serve as the production deployment target.

Answer

Vertex AI for prototyping, then Google AI Studio for production

Answer

Cloud Natural Language API for prototyping, then Vertex AI for production

Answer

Gemini for Google Workspace for both prototyping and production

Question 5

A media company wants to build a generative AI application that answers questions using its proprietary video transcripts. They need the model to cite specific transcript segments and avoid hallucinating content not present in those transcripts. Which Google Cloud approach best meets these requirements?

Accepted Answer

Use Vertex AI Search with grounded generation over the transcript data store. Grounded generation with Vertex AI Search retrieves relevant transcript passages and conditions the model's answer on them, which both limits hallucination and enables citations to specific segments. Fine-tuning, prompt-only instructions, or high temperature do not provide the retrieval and attribution needed for factual, source-backed answers.

Answer

Deploy Gemini with a system instruction to answer only from provided content, without retrieval

Answer

Increase the model temperature and include the entire transcript corpus in every prompt

Answer

Fine-tune a Gemini model on the full transcript corpus and deploy it to an endpoint

Question 6

A support team uses a generative AI assistant to answer customer questions. Agents report that the answers are often too long and include unnecessary background. The team wants the responses to be concise and directly address the question. Which prompt adjustment is most appropriate?

Accepted Answer

Tell the model to answer in two sentences or fewer and to skip background unless the customer asks for it.. The most appropriate adjustment is an explicit prompt constraint that limits answer length and omits background unless requested. Verbosity is a content and instruction-following issue, so a clear directive in the prompt is the right control. Sampling parameters like temperature and top-p do not directly enforce conciseness, and asking for more detail would worsen the problem.

Answer

Add a system instruction that the model should always provide a detailed explanation for every answer.

Answer

Lower the model's temperature to zero so answers become deterministic.

Answer

Increase the top-p value so the model considers more possible words.

Question 7

A financial services firm is evaluating Google Cloud generative AI offerings for an internal knowledge assistant. They need capabilities for controlling access to model endpoints and for monitoring model usage and safety signals. (Choose two.)

Accepted Answer

Cloud Monitoring dashboards and alerting for Vertex AI metrics. IAM roles and policies restrict and grant access to Vertex AI endpoints and related resources, satisfying the access-control requirement. Cloud Monitoring dashboards and alerts surface usage, latency, and error signals for those endpoints, satisfying the monitoring requirement. The remaining options address model export, feature serving, and storage lifecycle, none of which cover access control or operational monitoring.

Answer

BigQuery ML model export to Cloud Storage

Answer

Cloud Storage bucket lifecycle rules on training data

Answer

Vertex AI Feature Store for online serving

Question 8

A marketing team uses a Gemini model through the Vertex AI API to draft campaign copy. The drafts are creative but frequently wander off topic and include unsupported claims. The team wants a low-effort improvement before considering any model customization. Which action should they take first?

Accepted Answer

Rewrite the prompt to include a clear role, the target audience, explicit constraints, and a required output format.. Prompt engineering with an explicit role, audience, constraints, and output format is the fastest and cheapest way to align model output with a brief. Fine-tuning, larger models, and higher temperature all add cost or increase variance without addressing the underlying ambiguity in the instructions.

Answer

Raise the temperature so the model explores more ideas and eventually lands on better copy.

Answer

Switch the model to a larger version with more parameters to improve instruction following.

Answer

Create a supervised fine-tuning job with a few hundred labeled campaign examples.

Question 9

A financial services firm is deploying a generative AI model to answer customer questions about investment products. They want to ensure the model's responses are compliant with regulations and do not provide personalized financial advice. Which TWO techniques can help achieve this? (Choose two.)

Accepted Answer

Fine-tune the model on a dataset of compliant customer interactions.. Combining a system prompt with explicit instructions and fine-tuning on compliant interactions provides both immediate guidance and long-term behavioral shaping. The system prompt ensures the model follows rules in real time, while fine-tuning helps the model internalize compliance patterns, making responses more reliably safe. Together, they address the need for regulatory adherence and avoidance of personalized advice.

Answer

Deploy the model with a lower token limit to keep responses short and less detailed.

Answer

Use a content filter to block any response that contains words like 'recommend' or 'should invest'.

Answer

Increase the model's temperature to make responses more varied and less likely to give direct advice.

Question 10

A logistics company built a Gemini-powered assistant that answers driver questions about routes and hours-of-service rules. The assistant performs well on common questions but produces fabricated regulatory citations when asked about rare edge cases. The team has a curated set of correct answers for these edge cases and wants the model to adopt that behavior reliably. Which approach best fits?

Accepted Answer

Apply supervised fine-tuning on the curated edge-case examples to teach the desired response pattern.. Supervised fine-tuning uses the curated input-output pairs to adjust the model's weights so it reliably reproduces correct edge-case answers, which is the intended use of labeled examples. Prompt stuffing, sampling controls, and larger context windows do not durably change behavior for rare inputs.

Answer

Lower the topP value so the model only considers the most likely tokens and avoids fabricating citations.

Answer

Add the curated answers to the system instruction and rely on the model to generalize.

Answer

Increase the model's context window by switching to a long-context variant and pasting the full regulations.

Question 11

A marketing team uses a generative AI model to create short social media posts from long product briefs. The posts frequently include made-up statistics and product claims that are not in the brief. The team wants to reduce these unsupported statements without retraining the model. Which technique should they apply first?

Accepted Answer

Add grounding instructions that require the model to use only facts from the provided brief and to say when information is missing.. Grounding the prompt in the provided brief is the most direct way to stop unsupported statistics and claims. It tells the model exactly which facts are allowed and how to respond when a detail is missing. This approach avoids retraining costs and can be applied immediately, making it the right first step for the marketing team.

Answer

Reduce the maximum output token limit so the model cannot include extra details.

Answer

Increase the model's temperature setting to encourage more creative output.

Answer

Fine-tune the model on a dataset of approved social media posts.

Question 12

A healthcare analytics team uses a Gemini model to generate patient-friendly discharge summaries from clinical notes. Clinicians report that summaries occasionally omit critical follow-up instructions. The team must improve recall of these instructions without retraining the model. Which two techniques should they apply? (Choose two.)

Accepted Answer

Add an explicit instruction to extract all follow-up actions into a mandatory checklist section of the summary.. Few-shot examples and an explicit mandatory checklist section both shape the model toward systematically extracting follow-up actions, improving recall without retraining. Temperature changes, token limits, and deterministic decoding affect randomness or length, not whether required clinical content is captured in the output.

Answer

Reduce the maximum output tokens to force the model to be more concise and include only essentials.

Answer

Increase the temperature so the model produces more varied summaries and captures more details.

Answer

Enable deterministic decoding and assume the model will then always include every instruction.

Question 13

A media company wants to generate short video clips from text prompts for social media ads. The creative team needs a managed Google Cloud service that produces video from descriptive prompts and offers controls for aspect ratio and duration, without managing GPU infrastructure. Which offering should they use?

Accepted Answer

Vertex AI Veo. The team needs prompt-driven video generation delivered as a managed service, which is what Veo on Vertex AI provides, including controls over clip attributes and no infrastructure management. Imagen produces still images, Chirp targets speech, and Cloud Vision API analyzes existing images, so none of the alternatives can generate video content for the social ad campaign.

Answer

Vertex AI Chirp

Answer

Vertex AI Imagen

Answer

Cloud Vision API

Question 14

A product team at a retail company is using a foundation model on Vertex AI to generate short marketing taglines for new products. They find the outputs are often too long and sometimes include extra commentary. They want to constrain the model to produce only a single concise tagline. Which parameter should they adjust?

Accepted Answer

Max output tokens. The max output tokens parameter caps the total number of tokens generated, which directly enforces a limit on response length. By setting it appropriately, the team can ensure the model produces only a short tagline and cannot continue with extra commentary. Other parameters like temperature, top-k, and top-p affect randomness and diversity, not length.

Answer

Temperature

Answer

Top-p

Answer

Top-k

Question 15

A team is building a customer support assistant on Vertex AI using a foundation model. They notice the model occasionally invents policy details that don't exist in their internal documentation. They want the model to ground its answers in the company's knowledge base and provide citations. Which approach should they use?

Accepted Answer

Use Retrieval-Augmented Generation (RAG) with Vertex AI Search to retrieve relevant documents and include them in the prompt.. Retrieval-Augmented Generation retrieves relevant passages from an authoritative knowledge base and supplies them as context, so the model's answer is conditioned on real policy text rather than parametric memory. With Vertex AI Search, the system can also surface citations, letting users verify sources. This combination directly reduces fabricated policy details while preserving natural language fluency.

Answer

Reduce the max output tokens to limit the length of the model's responses.

Answer

Increase the temperature parameter to encourage more creative responses.

Answer

Fine-tune the model on a dataset of general customer service conversations.

Question 16

A financial analyst is using a large language model to generate executive summaries from lengthy earnings call transcripts. The summaries often miss key financial figures and include irrelevant details. Which technique should be used to improve the relevance and accuracy of the summaries?

Accepted Answer

Use few-shot prompting with examples of well-structured summaries.. Few-shot prompting is a powerful technique to guide the model's output by providing examples. In this scenario, showing the model examples of summaries that include key financial figures and exclude irrelevant details helps it learn the desired pattern, thereby improving relevance and accuracy.

Answer

Increase the maximum output token limit.

Answer

Fine-tune the model on a large corpus of general text.

Answer

Increase the model's temperature setting.

Question 17

A hospital network wants patients to describe symptoms in a mobile app and receive immediate guidance, but the clinical knowledge base changes weekly and the network must be able to update answers without retraining a model. They also need the assistant to escalate to a nurse when confidence is low. Which Google Cloud approach best meets these requirements?

Accepted Answer

Build a Vertex AI Agent Builder agent grounded on a frequently refreshed Vertex AI Search data store, with a tool that triggers nurse escalation. The requirements pair a knowledge base that changes frequently with a need to escalate to a human, which maps to an agent grounded on a refreshable data store plus a tool for escalation. Updating the search data store keeps content current without retraining, while function calling handles the nurse handoff. Fine-tuning, giant prompts, or self-managed open models fail one or both constraints.

Answer

Fine-tune a Gemini model on the clinical knowledge base each week and deploy it to a Vertex AI endpoint

Answer

Deploy an open model from Vertex AI Model Garden and manage the serving infrastructure on Google Kubernetes Engine

Answer

Use the Gemini API directly with a long system prompt containing the entire clinical knowledge base

Question 18

A customer support team uses a foundation model via the Gemini API to answer billing questions. Responses are often vague and sometimes omit required steps. The team wants to improve output quality without fine-tuning the model. Which approach should they use?

Accepted Answer

Use prompt engineering with few-shot examples and a clear output format that lists the required steps in order.. Prompt engineering with few-shot examples and an enforced output format is the fastest way to raise quality without changing the model weights. Examples demonstrate the expected depth, and a structured template guarantees that required steps appear. Sampling parameters alter randomness, and token limits alter length, but neither supplies missing procedural knowledge or enforces completeness.

Answer

Lower the top-p value so the model only considers the most probable tokens, which should make the answers more accurate.

Answer

Increase the maximum output tokens so the model has more room to produce a longer, more complete answer.

Answer

Raise the temperature setting so the model can explore more possible answers and include more detail in its responses.

Question 19

A software company is using a large language model to generate code snippets from natural language descriptions. The generated code often has syntax errors and does not follow the company's coding standards. Which approach is most effective to improve the quality of the generated code?

Accepted Answer

Fine-tune the model on a dataset of code that adheres to the company's coding standards.. Fine-tuning on a dataset of code that adheres to the company's coding standards is the most effective way to improve the quality of generated code. It directly teaches the model the desired syntax, style, and patterns, reducing syntax errors and ensuring compliance with standards.

Answer

Add a prompt instruction to 'write clean code'.

Answer

Increase the temperature to generate more diverse code solutions.

Answer

Use a larger model with more parameters without fine-tuning.

Question 20

A software vendor embeds Gemini in its SaaS product and needs each tenant's prompts and outputs isolated so that one tenant's data never appears in another tenant's responses. The vendor also wants usage tracked per tenant for billing. Which design should the team implement on Google Cloud?

Accepted Answer

A separate Google Cloud project per tenant with its own Vertex AI resources and service accounts, plus labels or billing export to attribute usage per tenant.. Strong tenant isolation comes from separating identities and resources, which separate Google Cloud projects per tenant provide, backed by distinct service accounts and Vertex AI resources. Labels and billing export then make consumption attributable per tenant. Prompt-level tagging, shared credentials, or network-only controls do not create the data or identity boundaries multi-tenant isolation requires.

Answer

One project with per-tenant data stores and a shared service account whose credentials are distributed to all tenant workloads.

Answer

A single shared Google Cloud project and one Gemini endpoint, with tenant identifiers included in the prompt text and per-tenant reporting derived from exported logs.

Answer

One project with VPC Service Controls and per-tenant Cloud Armor rules, without separate identities or data stores for tenants.