2.5 Configuring Prebuilt Solutions or APIs with Agent Platform PROFESSIONAL-CLOUD-ARCHITECT Practice Questions (Page 1)

Questions 1–5

1. 1foundation · easy

   Which Gemini Enterprise feature allows users to create custom AI assistants that can interact with enterprise data and perform tasks?

   • AGemini for Google Cloud
   • BVertex AI
   • CAI Agents
   • DNotebookLM
   Reveal Solution Discussion Select an answer first
2. 2expert · hard

   A company needs to build a system that analyzes customer support chat logs to detect frustration and escalate to a human agent. The chat logs are text, and the company has a limited budget and needs a solution that can be deployed quickly. They are considering models from Model Garden. Which type of model would be the most appropriate, and what is the key trade-off they should consider?

   • AA video analysis model; it can analyze the chat logs if they are converted to video.
   • BA custom model built from scratch; it will be perfectly accurate but requires significant time and data.
   • CA pre-trained sentiment analysis model; it is fast to deploy but may not be perfectly tuned to the company's specific support context.
   • DA fine-tuned model; it will be perfectly accurate but requires a large labeled dataset.
   Reveal Solution Discussion Select an answer first
4. 3foundation · easy

   What is Model Garden in Google Cloud?

   • AA database for storing trained model artifacts
   • BA curated collection of AI models and foundation models
   • CA tool for training custom machine learning models from scratch
   • DA service for deploying models to edge devices
   Reveal Solution Discussion Select an answer first
5. 4application · medium

   A developer has selected a pre-trained model from Model Garden and wants to integrate it into a serverless application on Google Cloud. What is the most straightforward way to call the model from the application?

   • ARe-train the model within the serverless function before each call.
   • BDeploy the model to a dedicated VM and have the serverless function connect to it via a VPN.
   • CDownload the model weights and load them into the serverless function's memory.
   • DUse the model's endpoint via a REST API call from the application code.
   Reveal Solution Discussion Select an answer first
6. 5application · medium

   A company's IT helpdesk receives many repetitive requests for password resets. They want to use an AI Agent to automate the initial response and resolution. What is the most important consideration when designing this AI Agent?

   • AMake the agent as fast as possible, even if it means skipping verification steps.
   • BUse the agent to send a password reset link to the email address on file without any further checks.
   • CEnsure the agent can securely verify the user's identity before performing a password reset.
   • DHave the agent respond to all requests with a generic message asking the user to call the helpdesk.
   Reveal Solution Discussion Select an answer first