AI-103 Practice Test 4 — 25 Questions

Question 1

You are troubleshooting a multi-agent workflow in Azure AI Foundry where an agent fails to invoke an external API tool correctly. Which THREE debugging steps should you perform? (Choose THREE)

Accepted Answer

Test network connectivity and authentication credentials between the agent service and the external API endpoint. Inspecting tool definition JSON schemas, reviewing model output tool call arguments, and verifying API endpoint connectivity are key troubleshooting steps.

Answer

Restart the client web browser session

Answer

Increase the Azure DNS TTL value for domain resolution

Question 2

You are evaluating generative AI model outputs in Azure AI Foundry. Which TWO automated evaluation metrics assess the quality and correctness of RAG responses against source data? (Choose TWO)

Accepted Answer

Groundedness (measuring adherence to source context). Groundedness and relevance are core Azure AI Foundry automated evaluation metrics used to assess RAG application quality.

Answer

Virtual network gateway throughput

Answer

Storage account transaction count per hour

Answer

CPU utilization percentage of the Azure App Service plan

Question 3

You are designing a secure RAG retrieval pipeline in Azure AI Search. Which THREE indexing or search configurations are essential when handling confidential documents? (Choose THREE)

Accepted Answer

Encryption at rest using Customer-Managed Keys (CMK). Security trimming (filters), encrypted indexes, and restricted network access via private endpoints are essential for confidential RAG pipelines.

Answer

Disabling all authentication to allow open guest queries

Answer

Storing unencrypted search indexes in public Azure blob containers

Question 4

When designing an Azure AI Foundry project architecture, which TWO resources are typically linked to the Azure AI Hub to enable full model deployment and search capabilities? (Choose TWO)

Accepted Answer

Azure AI Search service. Azure OpenAI and Azure AI Search are core resources linked to Azure AI Hub workspaces for model hosting and RAG search indexer capabilities.

Answer

Azure SQL Database relational server

Answer

Azure Stream Analytics job cluster

Answer

Power BI embedded analytics workspace

Question 5

When designing a Responsible AI evaluation workflow in Azure AI Foundry, which TWO metrics are specifically used to assess content safety and potential harms? (Choose two.)

Accepted Answer

Hate speech and unfairness detection. Violence, sexual, hate, and self-harm detection along with protected material detection are core safety metrics in Azure AI Foundry.

Answer

Word count frequency

Answer

BLEU machine translation score

Answer

Model parameter count

Question 6

You are implementing responsible AI practices in Azure AI Foundry for a public-facing chatbot. Which THREE safety evaluations or controls should you implement to protect users and the organization? (Choose THREE)

Accepted Answer

Protected material detection for source code and text copyright verification. Protected material detection, content safety filtering for harmful categories, and jailbreak/prompt shielding defenses are core responsible AI controls in Azure AI.

Answer

Azure SQL database transparent data encryption (TDE)

Answer

Azure DDoS network layer volumetric scrubbing

Question 7

You are deploying an LLM in Azure AI Foundry and need to monitor for toxicity and jailbreak attempts in real-time. Which feature should you enable?

Accepted Answer

Azure AI Content Safety. Azure AI Content Safety is the integrated service for monitoring model inputs and outputs for harmful content.

Answer

Azure Policy

Answer

Microsoft Purview Compliance Manager

Answer

Azure Monitor Log Analytics

Question 8

An agent is prone to 'hallucinations' when responding to specific niche queries. What is the most effective way to reduce these hallucinations using RAG?

Accepted Answer

Set the grounding threshold to a higher value in the agent configuration.. Increasing the 'top_k' or relevance threshold ensures the retrieved data is highly specific to the query, reducing ambiguity.

Answer

Retrain the base LLM on the niche dataset.

Answer

Increase the system prompt length.

Answer

Lower the temperature setting to 0.0.

Question 9

You are using Azure AI Foundry to deploy an agent. You want to ensure that the agent only uses approved data sources. Which feature should you configure?

Accepted Answer

Grounding data configuration in the Agent settings.. Grounding with specific data sources restricts the agent to provided data, preventing the use of external knowledge.

Answer

Prompt engineering for 'Do not use external info'.

Answer

Model temperature control.

Answer

Content Safety policy filters.

Question 10

An agent is designed to summarize meeting transcripts but keeps including internal meeting notes in the final summary. What is the most effective way to address this?

Accepted Answer

Update the system prompt to clarify which sections of the transcript to include.. Modifying the system prompt to explicitly define the inclusion/exclusion criteria for the summary is the most effective way to control agent output.

Answer

Lower the temperature.

Answer

Switch to a different model.

Answer

Add a 'stop_sequence' to the API call.

Question 11

You notice your Azure AI agent is producing offensive content. You need to configure a guardrail that intercepts the model output before it is sent to the user. Which service should you integrate?

Accepted Answer

Azure AI Content Safety. Azure AI Content Safety is the standard guardrail for filtering model output.

Answer

Azure AI Search Filters

Answer

Prompt Template Validation

Answer

Azure Monitor

Question 12

What is the primary benefit of using the Model Catalog in Azure AI Foundry?

Accepted Answer

Centralized discovery and deployment of models. The Model Catalog provides a centralized location to discover and deploy models.

Answer

Automated prompt engineering

Answer

Network security configuration

Answer

Automatic data labeling

Question 13

You are troubleshooting an Azure OpenAI deployment in Azure AI Foundry where agents are failing to complete tasks due to excessive timeouts. Which THREE diagnostic steps should you take? (Choose three.)

Accepted Answer

Check if the tokens-per-minute (TPM) quota is throttling requests, causing queued delays.. Troubleshooting agent timeouts involves examining token limits, network latency, and model complexity.

Answer

Double the model temperature to 2.0 to accelerate reasoning speed.

Answer

Delete all vector indices to force local memory processing.

Question 14

You are designing an agentic workflow where an agent needs to maintain conversational context and remember user preferences across multiple disparate sessions over several weeks. Which architectural component should you implement?

Accepted Answer

A persistent semantic memory store integrated with the agent's context pipeline.. Long-term memory implemented via vector stores or external databases allows agents to retrieve historical interactions across sessions.

Answer

Increasing the model context window to 1,000,000 tokens for every request.

Answer

Using a stateless function with zero memory persistence.

Answer

Storing the entire conversation history in a local client-side cookie.

Question 15

You are configuring an agent in Azure AI Foundry that needs to perform mathematical calculations accurately. Large language models often make arithmetic errors. How should you equip the agent to handle calculations reliably?

Accepted Answer

Provide a code interpreter tool so the model can write and execute Python code for calculations.. Giving the agent a code interpreter or calculator tool allows it to execute calculations deterministically rather than relying on internal probabilistic math.

Answer

Set the model temperature to 0.0 and ask it to guess the result.

Answer

Increase the presence penalty to force new tokens.

Answer

Instruct the model in the system prompt to double-check its arithmetic mentally.

Question 16

You are evaluating a generative AI application in Azure AI Foundry using automated metrics. You need to measure how well the generated answers are supported by the retrieved source documents. Which evaluation metric should you use?

Accepted Answer

Groundedness. Groundedness measures the extent to which the generated text is derived solely from the provided source context, preventing hallucinations.

Answer

Word Error Rate (WER)

Answer

BLEU score

Answer

Perplexity

Question 17

An agent is performing poorly on multi-hop reasoning tasks. What architectural pattern should you implement to improve performance?

Accepted Answer

Use a Chain-of-Thought prompting strategy in the agent's system prompt.. Chain-of-thought prompting forces the model to break down complex queries into intermediate logical steps.

Answer

Increase the temperature to 1.0.

Answer

Switch to a smaller model to increase speed.

Answer

Decrease the max_tokens parameter.

Question 18

Which TWO of the following are valid methods to improve the quality of retrieved data in a RAG system?

Accepted Answer

Implementing a hybrid search strategy. Better chunking strategies and hybrid search improve retrieval quality.

Answer

Reducing the number of vector dimensions

Answer

Changing the chat history session length

Answer

Increasing the number of tokens in the prompt

Question 19

You are designing an advanced agentic solution in Azure AI Foundry that requires tool integration and state management. Which TWO practices should you implement to ensure robust agent reliability and security? (Choose two)

Accepted Answer

Implement human-in-the-loop confirmation steps before the agent executes high-impact or destructive transactional tool calls.. Validating tool inputs and implementing human-in-the-loop validation for sensitive actions are critical best practices for agent reliability and security.

Answer

Disable all logging and telemetry to minimize latency during agent execution loops.

Answer

Grant the agent unrestricted administrative permissions to all corporate databases to maximize autonomous problem-solving.

Answer

Rely solely on the model's internal memory to maintain multi-session conversation state without external persistence.

Question 20

You are designing a generative AI application that processes user uploaded documents. You must ensure that Personally Identifiable Information (PII) such as Social Security numbers are automatically redacted before the text is sent to the Azure OpenAI model. Which service should you integrate into your pipeline?

Accepted Answer

Azure AI Language service (PII detection and redaction). Azure AI Language provides dedicated features for PII detection and redaction (named entity recognition for PII).

Answer

Azure AI Translator

Answer

Azure AI Vision OCR

Answer

Azure Monitor Application Insights