1Z0-1127 Practice Test 13 — 15 Questions

Question 1

An architect is designing a GenAI solution for document summarization that must meet GDPR compliance. The data should not leave the EU. OCI GenAI models are available in Frankfurt, London, and Paris. Which is the best approach?

Accepted Answer

Deploy a dedicated AI cluster in Frankfurt and upload data to Object Storage in Frankfurt.. Option D is correct because deploying a dedicated AI cluster in Frankfurt ensures data stays within the EU region and does not traverse to other regions for inference.

Answer

Use the managed serving endpoint in Frankfurt.

Answer

Use the playground in any EU region.

Answer

Use a pre-trained model from OCI's catalog.

Question 2

Refer to the exhibit. The API Gateway fails to invoke the Generative AI service. What is the most likely missing configuration?

Accepted Answer

The API Gateway is not in the same VCN as the service's private endpoint.. Since the service endpoint is private, the API Gateway must be in the same VCN to have network connectivity.

Answer

The API Gateway does not have an internet gateway.

Answer

The JWT token is expired.

Answer

The Generative AI model is not deployed.

Question 3

Which TWO actions are required to enable a user to access OCI Generative AI service?

Accepted Answer

Subscribe to the GenAI service in the OCI Console.. Options A and B are correct. The user must have an IAM policy granting access to the GenAI service, and the service must be subscribed (enabled) in the tenancy. The other options are not prerequisites.

Answer

Create a dedicated AI cluster.

Answer

Enable the GenAI service in the region's service limits.

Answer

Install the OCI SDK.

Question 4

Which THREE factors should be considered when choosing a model for a summarization task using OCI Generative AI?

Accepted Answer

Number of parameters.. Options A, C, and D are correct. Context window size (A) determines how much text the model can process at once. Number of parameters (C) affects capability and cost. Supported languages (D) ensures the model can handle the input language. Option B is wrong because model training data cut-off date is less critical for summarization. Option E is wrong because inference endpoint location is about deployment, not model selection.

Answer

Inference endpoint location.

Answer

Model training data cut-off date.

Question 5

A company wants to use OCI Generative AI service to generate email summaries for customer support. They need to ensure low latency and data residency in Frankfurt. What should they use?

Accepted Answer

Use the managed serving endpoint in Frankfurt region.. Option A is correct because the managed serving endpoint is available in Frankfurt region, providing low latency without the need for a dedicated cluster. Dedicated AI cluster is not necessary for latency and data residency if the managed endpoint is in the same region. The playground is for testing, and Quick Actions are for pre-built models.

Answer

Use the playground in US region.

Answer

Use OCI AI Quick Actions.

Answer

Create a dedicated AI cluster in Frankfurt.

Question 6

Which THREE are valid ways to interact with OCI Generative AI?

Accepted Answer

OCI REST API.. Options A, B, and C are correct. The OCI Console Playground, CLI, and REST API are all direct interfaces. The others are not standard ways to invoke the service.

Answer

OCI Mobile App.

Answer

OCI Data Science Notebooks.

Question 7

A company uses OCI Generative AI service for customer support summarization. They notice the model frequently misses key details and generates hallucinations. What should they do first?

Accepted Answer

Adjust the prompt to be more specific and include few-shot examples.. Option C is correct because improving prompt engineering with specific instructions and few-shot examples reduces hallucinations and improves accuracy. Option A is wrong because increasing temperature increases randomness, making hallucinations worse. Option B is wrong because switching models is a more drastic step that may not address the root cause. Option D is wrong because increasing max tokens does not improve accuracy.

Answer

Increase the temperature parameter.

Answer

Use a different base model.

Answer

Increase the max tokens.

Question 8

Which TWO are benefits of using OCI Generative AI service's dedicated AI cluster?

Accepted Answer

Ability to fine-tune models on custom data.. Options A and B are correct. A dedicated AI cluster allows fine-tuning on custom data and offers lower latency compared to serverless inference. Option C is wrong because fine-tuning requires training data. Option D is wrong because dedicated clusters have fixed capacity and do not auto-scale. Option E is wrong because content filtering is not a specific benefit of dedicated clusters.

Answer

Automatic scaling to handle large workloads.

Answer

Built-in content filtering for all outputs.

Answer

No need to provide any training data.

Question 9

A developer is building a RAG chatbot for an internal knowledge base. To ensure the system retrieves the most relevant chunks, what is the best practice for chunking?

Accepted Answer

Use semantic chunking with overlap. Semantic chunking with overlap ensures that context is preserved and retrieval is more accurate by avoiding splits in meaningful content.

Answer

Use very small chunks

Answer

Use fixed-size chunks without overlap

Answer

Use random-sized chunks

Question 10

A company uses OCI Generative AI to create embeddings for a vector search. They notice high latency in search queries. What is one possible optimization?

Accepted Answer

Use exact search for better accuracy. Approximate nearest neighbor (ANN) search trades a slight reduction in accuracy for a significant speedup, addressing high latency.

Answer

Decrease batch size for embedding creation

Answer

Use approximate nearest neighbor (ANN) search

Answer

Increase the embedding dimension

Question 11

An application uses RAG to answer customer queries, but answers are often incomplete because the retrieved chunks do not contain full context. Which adjustment should the developer make?

Accepted Answer

Increase the number of retrieved chunks. Increasing the number of retrieved chunks gives the model more contextual information, leading to more complete answers.

Answer

Use a different embedding model

Answer

Increase chunk overlap

Answer

Decrease chunk size

Question 12

A team uses OCI OpenSearch as a vector database for RAG. Some queries return no results despite relevant documents being indexed. What is a likely cause?

Accepted Answer

The number of candidates (k) is set too low. If k (the number of nearest neighbors to return) is set too low, the search may not find any documents within the top k, returning no results.

Answer

The vector index is not refreshed after adding documents

Answer

The query text is too long for the embedding model

Answer

The embedding model is incompatible with the document language

Question 13

An enterprise RAG system must ensure that retrieved data comes only from authorized sources. Which OCI feature should be used to enforce this?

Accepted Answer

OCI IAM policies for the vector database. OCI IAM policies allow fine-grained access control to resources, ensuring only authorized users or services can retrieve data from the vector database.

Answer

Data encryption at rest

Answer

Network security groups

Answer

Resource quotas

Question 14

A team fine-tunes an embedding model for a legal document RAG system but observes low retrieval recall. Which technique is most likely to improve recall?

Accepted Answer

Use hard negative mining during training. Hard negative mining exposes the model to challenging negatives during training, which sharpens the embedding space and improves recall.

Answer

Use a smaller batch size

Answer

Reduce the learning rate

Answer

Increase the number of fine-tuning epochs

Question 15

A developer uses OCI Generative AI with a custom OCI OpenSearch vector store. The text generation model sometimes hallucinates facts not in the retrieved documents. What is the most effective mitigation?

Accepted Answer

Use prompt engineering to instruct the model to stick to the provided context. Prompt engineering with clear instructions to use only the provided context is a direct and effective way to reduce hallucination.

Answer

Use a larger retrieval chunk size

Answer

Increase the temperature

Answer

Decrease the maximum token length