At Velmora Systems, You are building a flow-based repair-booking agent in CX Agent Studio.

GCP Google Cloud Professional Agentic Architect Practice Questions At Velmora Systems, You are building a flow-based repair-booking agent in CX Agent Studio.

GCP Google Cloud Professional Agentic Architect Practice Questions

0% Completed

Premium Set: 1 [ + Detailed Explanations] Free Preview

At Pollsafe Software, Antigravity reviews an ADK Python polling tool.

At Duranta Review Systems, An ADK Python application's administrative endpoint updates a review flag outside an agent invocation:

Reading

At Flashline Manufacturing, A Gemini Enterprise workflow sends short packaging-line videos to a Gemini preprocessing step before indexing inspection observations.

Reading

At Zephora Analytics, A production RAG service stores document vectors generated by gemini-embedding-001.

Reading

At Kiventra Labs, A team uses Antigravity across several application repositories. Developers repeatedly ask it to follow the same API-migration procedure, including a checklist, example patches, and a validation script.

Reading

At Caldris Research, An agent identity has IAM permission to read objects in two projects: research and finance.

Reading

At Norvexa Research, A RAG assistant answers questions about laboratory equipment.

Reading

At Throughputia AI, An agent's Gemini requests use a model, project, and region covered by an active Provisioned Throughput purchase.

Reading

At Cresovia Research, An assistant sends the same 60,000-token reference handbook with hundreds of different Gemini requests each hour.

Reading

At Silverpine Systems, An ADK coordinator reads a configuration snapshot from three independent replica agents.

Reading

At Orbelis Integration, A custom ADK connector discovers tools from an MCP server.

Reading

At Quenara Observability, An agent service exports traces through an OpenTelemetry collector to Cloud Trace.

Reading

At Orbinex Solutions, An ADK fulfillment assistant uses InventoryCheck and TransitCheck agents to query separate systems.

Reading

At Morrowbyte Labs, An agent platform is adding a custom stdio MCP client for servers implementing protocol version 2026-07-28.

Reading

At Repairhaven Technologies, A maintenance agent uses Cloud SQL for PostgreSQL as its authoritative repair catalog.

Reading

At Corvexis AI, A Python service implements manual function calling with Gemini 3.

Reading

At Pinforge Software, Antigravity maintains a Python agent project whose clean CI builds occasionally install different transitive dependency versions from the same commit.

Reading

At Keytone Membership Services, A CX Agent Studio telephone agent collects a variable-length membership code in a form parameter using the keypad.

Reading

At Triarc Cloudworks, A Cloud Run agent calls a tool adapter, which uses an SDK to perform a read-only supplier lookup.

Reading

At Nuvexa Technologies, A small team packages an ADK assistant as a standard HTTP container.

Reading

At Norvexa Logistics, An ADK fulfillment agent may call get_price and check_stock in either order, but must call reserve_stock exactly once after both succeed.

Reading

At Dispatchora Operations, An ADK maintenance workflow commits a work order to Cloud SQL for PostgreSQL, then sends an event to an external dispatcher.

Reading

At Brighthaven Analytics, An unattended analysis agent runs in an on-premises environment and submits BigQuery jobs.

Reading

At Verdantis Consulting, An ADK Python assistant requests a compliance-document conversion that takes about 40 minutes.

Reading

At Livebridge Voice, A browser-based voice assistant uses the Gemini Developer API for direct Live API connections.

Reading

At Meralune Travel, An ADK event-planning assistant passes single-turn tests that explicitly include the event date.

Reading

At Projecta Secure, A support agent's export pipeline uses Sensitive Data Protection to detect confidential project codes.

Reading

At Arclune Technologies, An ADK agent calls an MCP tool to reserve a training room.

Reading

At Tervalon Industries, A manufacturer is preparing a Gemini Enterprise search app over manuals and restricted supplier agreements stored in Cloud Storage.

Reading

At Labelcrest Systems, Antigravity reviews a Python tool on Linux that prints a user-supplied label:

Reading

At Velmora Systems, You are building a flow-based repair-booking agent in CX Agent Studio.

Reading

At Traceguard Analytics, An agent's OpenTelemetry instrumentation attaches raw prompts and tool payloads to trace-span attributes.

Reading

At Quendrix Systems, A platform runs coding agents in disposable Pods on a shared GKE cluster.

Reading

At Trevanta Commerce, Antigravity is optimizing an agent's order-summary tool, which uses SQLAlchemy with Cloud SQL for PostgreSQL.

Reading

At Veridane Services, A CX Agent Studio assistant invokes a shared verification flow from several different business flows.

Reading

At Rankwell Research, A team evaluates the ranked passages supplied to its RAG agent.

Reading

At Brighthollow Systems, An engineering team connects Antigravity to a custom MCP server for a Cloud SQL for PostgreSQL reporting database.

Reading

At Diversa Procurement, A procurement assistant retrieves ten product candidates from Vector Search 1.0.

Reading

At Luminara Services, A CX Agent Studio page has these condition-only transition routes in the displayed order:

Reading

At Bluecrest Platforms, A team is configuring Agent Gateway egress for agents on Agent Runtime.

Reading

At Nexora Labs, An ADK research agent has an HTTP-fetch tool that checks the initial URL against approved public documentation hosts.

Reading

At Astrivex Systems, A team is adding the Agent Platform Gen AI evaluation service to CI for an agent handling delivery inquiries and damaged-goods claims.

Reading

At Luminara Engineering, A team uses Antigravity with Agents CLI to maintain an ADK project generated from an older scaffold.

Reading

At Noventra Archives, An agent team must annotate a large, one-time archive before loading it into a knowledge system.

Reading

At Velmora Labs, A museum uses a CX Agent Studio playbook to handle booking changes and cancellations.

Reading

At Elaris Technical Media, A technical-writing agent retrieves current product facts correctly, but repeatedly fails the company's specialized writing conventions.

Reading

At Regexa Secure Systems, Antigravity reviews a Python agent tool that validates untrusted text with re.fullmatch.

Reading

At Asterquill Learning, A learning assistant already stores each conversation reliably.

Reading

At Catalogica Search, A catalog-search agent uses gemini-embedding-001 for product descriptions and incoming search phrases such as "quiet portable compressor."

Reading

At Attestica Cloud, An agent team stores container images in Artifact Registry and deploys them to Cloud Run.

Reading

Premium Set: 2 [ + Detailed Explanations]

At Solmera Storage Services, An agent's backend downloads nonpublic Cloud Storage objects through the client library using a dedicated service account.

Reading

At Noventra Works, A service calls Gemini to classify incoming maintenance tickets.

Reading

At Aurelis Capacity Systems, An application uses Gemini function calling for a capacity-planning assistant.

Reading

At Praxilon Tools, Antigravity launches a custom MCP server using stdio.

Reading

At Vantorra Shipping, An agent on Cloud Run calls a third-party API to create shipping labels.

Reading

At Tavrion Services, An inspection agent on Cloud Run stores workflow progress in Firestore.

Reading

At Meridian Warranty Labs, A support assistant retrieves a newly approved warranty policy correctly, but its evaluation score falls because several golden answers still describe the superseded policy.

Reading

At Novalune Travel, A travel assistant calls an optional recommendation API that sometimes remains unavailable for several minutes.

Reading

At Monderra Systems, A team evaluates two agent versions with a custom simulator that returns recorded tool responses in sequence.

Reading

At Bucketshield Security, An organization-level IAM deny rule blocks Cloud Storage object reads for all principals.

Reading

At Londara Operations, A team builds a daily operations briefing in Gemini Enterprise's low-code Workflow Builder.

Reading

At Veltrixa Logistics, A CX Agent Studio delivery agent collects a shipping method. Customers say "express delivery" or "next-day shipping," but the fulfillment API requires the canonical value EXPRESS for either phrase.

Reading

At Merovex Labs, At its welcome page, a CX Agent Studio voice agent waits for a request.

Reading

At Publishara Enterprise, Antigravity has helped a team deploy an ADK assistant successfully to Agent Runtime.

Reading

At Formguard Web, Antigravity reviews a web interface that launches agent workflows through authenticated HTML form POST requests.

Reading

At Crestivo Software, A platform team has built an Antigravity IDE review skill, a specialist subagent, MCP server definitions, and lifecycle hooks.

Reading

At Mirrorcrest Services, A team wants to shadow a candidate account-service agent using copies of live requests while the baseline continues serving every user.

Reading

At Bookstate Labs, A booking agent is evaluated against isolated sandbox records.

Reading

At Twinforge Data, Two agent workers update different entries in an existing JSON manifest in Cloud Storage.

Reading

At Pelorin Procurement, A procurement agent can draft purchase orders and submit them through a tool.

Reading

At Cerynth Consulting, An ADK Python service supports consultants working with several client companies.

Reading

At ReportRelay Agents, An ADK Python coordinator already calls a report specialist through AgentTool.

Reading

At Phishguard Digital, An agent summarizes external troubleshooting pages and includes their links in its responses.

Reading

At Catalogis Data, Antigravity is optimizing an agent's in-memory catalog tool:

Reading

At Pirellex Operations, An ADK coordinator reserves equipment, then assigns a crew to that reservation, and finally requests a site permit.

Reading

At Elvaris Technologies, Different teams operate reusable agents and MCP servers across an organization.

Reading

At Norvanta Digital, Antigravity is reviewing a web assistant that inserts model-generated HTML into the conversation panel using innerHTML.

Reading

At Brinova Group, A company is connecting an ADK sales assistant to a third-party CRM.

Reading

At Auralis Systems, Antigravity proposes a refactor that reduces the CPU time of an invoice-calculation service.

Reading

At Deployora Labs, An MCP deployment-planning tool using protocol revision 2026-07-28 receives a request without a target region.

Reading

At Cresthaven Coaching, A Cloud Run assistant will compare two coaching strategies during a four-week online experiment.

Reading

At Solvanta Engineering, An Agent Search app indexes equipment manuals with product_line and revision metadata.

Reading

At Drainwell Kubernetes, During planned GKE rollouts, an agent Pod uses a required 10-second preStop hook.

Reading

At Atlasen Networks, A network-product assistant retrieves from Vector Search using only the latest user message.

Reading

At Caldris Engineering, Antigravity is helping optimize an ADK Python service.

Reading

At Twinwarehouse AI, A Gemini-based agent requests two lookup_stock function calls in one turn, one for warehouse East and one for warehouse West.

Reading

At Helion Forge, An ADK service on Agent Runtime calls a remote eligibility MCP tool.

Reading

At StereoScribe Support, A Gemini Enterprise ingestion workflow transcribes stereo support recordings with Speech-to-Text v1. Channel 1 always contains the employee and channel 2 the customer.

Reading

At Ardentis Labs, An ADK research assistant compacts history after every 20 conversation turns.

Reading

At Safepath Analytics, Antigravity reviews a Python tool that reads files selected by an agent from /srv/knowledge.

Reading

At Loopguard Desktop, A desktop coding assistant connects to a local MCP server over Streamable HTTP.

Reading

At Miradex Engineering, A platform team uses Antigravity to maintain internal services.

Reading

At Metricridge Platforms, An agent platform exports request counters and latency distributions to Cloud Monitoring with model, outcome, and session_id labels.

Reading

At Quantiforge Compute, A team must serve an approved open-weight model for its ADK assistant on one fixed GPU in GKE.

Reading

At Zenvora Cloudworks, An agent on Agent Runtime calls a registered external MCP server through Agent Gateway in Agent-to-Anywhere mode.

Reading

At Openroute Integrations, An ADK Python assistant must call a company's existing REST service.

Reading

At Partivane Systems, A CX Agent Studio agent must query an existing HTTPS parts API.

Reading

At Vocalune Interactive, A voice agent uses the Gemini Live API and buffers generated audio for browser playback.

Reading

At Statelessia Documents, An ADK Python workflow includes a stateless extraction agent.

Reading

At Solvanta Maintenance, A maintenance assistant retrieves 40 candidate passages from Vector Search 1.0 but can send only five to the model because of its context budget.

Reading

Premium Set: 3 [ + Detailed Explanations]

At Clockwise Labs, Antigravity is fixing a flaky token-expiration test. Production code reads the system clock directly.

Reading

At Pairwise Metrics, A candidate agent and the current production agent are evaluated on the same 2,000 representative cases, producing a success or failure outcome for each version on every case.

Reading

At Nexora Analytics, A coding agent performs an iterative Python data-analysis task.

Reading

At Forecastra Data, Antigravity reviews an agent tool that receives forecast records from independently operated suppliers and calls pickle.loads on the response bytes.

Reading

At BrindlePeak Systems, A Cloud Run agent service performs CPU-heavy document preprocessing in one application thread.

Reading

At Orivanta Labs, An MCP server receives a signed access token from an approved identity provider.

Reading

At Routehaven Services, A CX Agent Studio flow has four pages that should all handle the same two intents for changing language and requesting help.

Reading

At Corvanta Works, A repository contains Terraform configuration and unrelated application code.

Reading

At Fluxora Ingestion, Antigravity is optimizing an agent's ingestion tool on Cloud Run.

Reading

At Nexlume Search, A semantic-search agent uses Vector Search 1.0 with DOT_PRODUCT_DISTANCE and no feature normalization.

Reading

At Cancelpoint Coordination, An A2A coordinator has an active supplier task that is no longer needed because the user changed the project scope.

Reading

At Draventa Systems, An MCP proxy connects registered clients to a SaaS API using one shared upstream OAuth client ID.

Reading

At Merivox Services, An ADK service supports several markets.

Reading

At Omnitrace Agents, An ADK Python application has a coordinator and several nested specialist agents.

Reading

At Asterwyn Learning, A CX Agent Studio enrollment agent has collected a member number, email address, and course selection as required page form parameters.

Reading

At Ironvale Field Services, A field-service agent must answer new troubleshooting questions while an industrial site has no WAN connectivity for several hours.

Reading

At Teralune Systems, An application extracts standard form fields and analyzes conflicting policy clauses.

Reading

At Cinderpeak Security, Application teams manage their own Model Armor templates.

Reading

At Nexora Works, During a large refactoring task, an Antigravity agent needs a dependency specialist to inspect hundreds of manifests and report version conflicts.

Reading

At Talveron Design, An A2A coordinator tracks two tasks with one supplier agent in the same project context:

Reading

At Caldera Finch, A Gemini Enterprise knowledge assistant searches training videos stored in Cloud Storage.

Reading

At Velmora Analytics, A team creates six paraphrases of each source support case for agent evaluation.

Reading

At Temporis Agents, During one ADK invocation, a coordinator stores a large normalized tool response that two subagents need later in the same invocation.

Reading

At Progressa Imaging, An MCP client calls a long-running image-analysis tool.

Reading

At Dualpool Cloud, A Cloud Run agent's two independent lookup tools share one 20-thread executor.

Reading

At Myridian Policy Labs, A policy assistant retrieves small chunks from Vector Search 1.0. Top results identify a rule, but answers omit an exception elsewhere in the same section.

Reading

At Refinerya Documents, An ADK document agent should repeatedly critique and revise a draft until a checker determines that no actionable issues remain.

Reading

At Venuecraft Events, A CX Agent Studio event agent extracts a workshop name using a custom entity type.

Reading

At Quoralis AI, A production agent gets one invocation per user request; the application neither retries the task nor selects among alternative answers.

Reading

At Elvaris Research, An ADK Python incident-analysis agent repeatedly issues the same read-only search and receives unchanged results.

Reading

At Cachelume Systems, An ADK Python assistant repeatedly calls a read-only public catalog tool.

Reading

At Connexis Cloud, A Cloud Run agent opens up to five database connections per instance.

Reading

At Rubriva Quality Labs, A team is creating human labels for a nuanced agent-quality rubric.

Reading

At Evidencewell Research, A RAG assistant's task-success rate drops after a release that changes both retrieval and generation prompts.

Reading

At Orivanta Labs, An engineering coordinator must delegate design reviews to supplier-operated agents.

Reading

At Solvanta Parts, A replacement-parts assistant uses dense-vector retrieval in Vector Search 1.0.

Reading

At Thinkwell Classification, A classification agent uses a Gemini model with documented HIGH and LOW thinking levels.

Reading

At Docstream Engineering, A team uses Antigravity to build an ADK application against the current release.

Reading

At Memorix Labs, A new Memory Bank deployment creates and updates every memory through GenerateMemories.

Reading

At Valecroft Knowledge, A Gemini Enterprise assistant needs to search an external knowledge system.

Reading

At Meridian Finch, Antigravity reviews a report-download tool whose middleware authenticates each user correctly.

Reading

At Ninetynine Quality, A release gate requires an agent's true task-success rate to be at least 99%, supported by a one-sided 95% exact binomial lower confidence bound.

Reading

At Tandemridge Agents, An A2A coordinator receives these ordered artifact updates for one task.

Reading

At Pinecrest Codeworks, A team uses Antigravity to edit a TypeScript repository. After any file-writing tool completes, the workspace's formatter must run automatically so generated changes are normalized even when the agent forgets to request formatting.

Reading

At EightPeak Workspaces, Developers run Claude Code on Cloud Workstations.

Reading

At Birchstone Platforms, A platform team tests four agent configurations on the same 1,000 representative tasks.

Reading

At Handbookia Systems, A RAG ingestion service sends each long handbook as one input to gemini-embedding-001, then stores one vector per handbook.

Reading

At Vistalume Claims, A claims assistant receives a damage photo together with the user's written description.

Reading

At Calyvera Operations, A Gemini Enterprise workflow should prepare an internal triage brief when a new Jira Cloud issue with priority P1 is created.

Reading

At Quenaris Labs, An internal assistant streams generated text to browsers and sends a completed copy to Model Armor afterward.

Reading

Premium Set: 4 [ + Detailed Explanations]

At Orbinex Labs, Antigravity is reviewing a Python tool that looks up shipments in Cloud SQL for PostgreSQL using Psycopg.

Reading

At Entitlebridge Enterprise, An enterprise has a central entitlement service that makes authorization decisions from contract status, risk level, and temporary exceptions.

Reading

At Arborcrest Analytics, An agent's BigQuery tool runs single SELECT statements against native, non-clustered tables using on-demand pricing.

Reading

At Monitorwell AI, A production agent's prompts and model version have not changed, but the team is concerned that evolving user behavior may reduce response quality over time.

Reading

At Aurenix Searchworks, A retrieval team compares a Tree-AH Vector Search index with brute-force search over identical vectors using the same distance function and result count.

Reading

At Caldris Travel, An ADK trip-planning assistant sometimes needs a specialist agent to assess accessibility requirements.

Reading

At Keylance Integrations, An Agent Runtime tool calls a partner API that requires a long-lived API key.

Reading

At Inspectara Enterprise, A Gemini Enterprise team is preparing a new data store for short PDF inspection reports.

Reading

At Policyweave Systems, An agent is permitted to read payroll data and send external email in legitimate workflows.

Reading

At Quenvora Software, A team runs two Claude Code sessions on a Cloud Workstation.

Reading

At Novera Analytics, An evaluation pipeline uses a Gemini-based pairwise judge to compare two versions of a troubleshooting agent.

Reading

At Permivault AI, A RAG assistant caches answers per authenticated user, retaining every source document ID used to produce each answer.

Reading

At Lunexa Digital, An assistant blocks requests according to a Model Armor template. Reviewers find that its prompt-injection and jailbreak filter flags several benign requests as medium-confidence matches.

Reading

At Praxion Security, An agent's custom MCP adapter refreshes tool definitions from an authenticated supplier endpoint.

Reading

At Selvara Research, A RAG assistant should answer questions supported by an approved knowledge base and abstain when it lacks evidence.

Reading

At Auditcrest Platforms, A platform team has written new Agent Gateway IAM access policies for outbound agent traffic.

Reading

At Claimora Finance, A Gemini Enterprise Workflow Builder process receives a list of expense claims.

Reading

At Clearstone Compliance, A compliance assistant must answer questions that require combining a retention rule from one policy with an exception stored in a separate policy.

Reading

At Ironcrest Financial, An agent has legitimate IAM access to sensitive BigQuery and Cloud Storage resources in a regulated project.

Reading

At Northbridge Records, An ADK document-processing workflow must extract a record and validate it.

Reading

At Virelta Systems, A CX Agent Studio team edits a flow-based agent while its custom chat integration serves customers.

Reading

At Regalyn Market Intelligence, A market-monitoring agent answers questions about newly published public regulatory announcements.

Reading

At Safepath Case Systems, An agent invokes an external case-management MCP tool through Agent Gateway using non-streaming tools/call requests.

Reading

At Axioncrest Design, An ADK design assistant generates several revisions of a PNG floor plan during one conversation.

Reading

At Cendora Labs, A team uses Agents CLI with Antigravity to prepare Cloud Run deployment configuration from natural-language requirements.

Reading

At Talvion Works, A small team is building a custom Gemini assistant over supported PDF and text manuals in Cloud Storage.

Reading

At Veltrona Systems, A CX Agent Studio agent has a single multi-page onboarding flow.

Reading

At Vocalis Assist, A voice assistant uses the Gemini Live API with session resumption enabled.

Reading

At Percentra Platforms, An agent platform reports service-wide 95th-percentile latency by averaging the 95th percentiles reported by individual instances.

Reading

At Harborline Support, An ADK Python team has approved several complete customer-service conversations as regression cases. Each case includes multiple user turns, initial session state, expected tool use, and reference responses.

Reading

At Accessora Support, A CX Agent Studio handoff generator summarizes a support conversation using only $conversation as its source text.

Reading

At Myridian Data, An analytics team prepares tabular conversation records for evaluating a support agent.

Reading

At Runbookia Systems, An ADK integration uses MCP protocol revision 2026-07-28 and caches a runbook resource under a stable URI.

Reading

At Longlink Collaboration, A real-time collaboration agent requires each client to maintain one uninterrupted WebSocket connection for up to four hours.

Reading

At Frontara Platforms, External applications invoke an ADK agent on Agent Runtime through streamQuery with text-only requests.

Reading

At Northvale Compute, A GKE agent container loads a local model before its health endpoint can respond.

Reading

At Loadcrest Services, An agent service must handle new requests arriving at 50 requests per second even when responses slow down.

Reading

At Schemaforge Systems, A release pipeline evaluates an agent that must return JSON with three required fields, an approved status enum, an integer score from 0 through 100, and an identifier matching a fixed pattern.

Reading

At Keepalive Labs, Antigravity profiles an ADK Python tool that creates and closes a new httpx.AsyncClient for every lookup to the same HTTPS backend.

Reading

At Querybridge Analytics, An ADK analytics agent must inspect BigQuery datasets and execute supported BigQuery operations through MCP.

Reading

At Quickhaven Agents, An agent deployed on Agent Runtime receives only a few requests each hour.

Reading

At Shieldmere AI, An agent calls a remote MCP tool whose natural-language response can contain untrusted content from third-party documents.

Reading

At Brinexa Services, An Agent Search data store is populated nightly from a Cloud Storage manifest containing the complete authoritative set of current service bulletins.

Reading

At Productiva Search, A team is building a custom RAG service over JSON product records.

Reading

At Buildhaven Systems, Antigravity reviews an agent application's Dockerfile.

Reading

At Pinehaven Software, Antigravity is fixing a web assistant's mobile layout.

Reading

At Identara Security, An organization deploys many Agent Runtime agents that currently share one broad workload identity.

Reading

At Skillhaven Enterprise, An enterprise has developed approved agent skills containing instructions, executable code, and documentation.

Reading

At Longview Coordination, An A2A coordinator starts a supplier task that can run for several hours.

Reading

At Solvanta Logistics, Several Cloud Run agent services share a supplier API credential limited to 120 requests per minute across all clients.

Reading

Announcements Reviews Live Classes Course Info

Previous Next

GCP Google Cloud Professional Agentic Architect Practice Questions At Velmora Systems, You are building a flow-based repair-booking agent in CX Agent Studio.

GCP Google Cloud Professional Agentic Architect Practice Questions

Reset Course Progress?

Finish Course Early?

Printing Disabled