Connect to and consume Azure services (20–25%)

Microsoft Certified: Azure AI Cloud Developer Associate · 50 questions

• A city permits API must accept large document-summarization jobs without making residents wait on the HTTP response. Which Azure messaging pattern should the back end use?
• A 311 triage publisher must route the same incident message so housing and public-works AI workers each receive only their category. What Service Bus feature fits?
• RAG re-index messages keep failing after poison payloads reach the city’s embedding worker. Where should those failed Service Bus messages go for investigation?
• A clerk portal sometimes retries the same AI job submit, risking double token spend. How should Service Bus protect the city’s summarization queue?
• A resident chatbot must keep multi-step agent tool calls ordered for one conversation. Which Service Bus capability enforces that ordering?
• Container Apps workers summarizing permit packets are thrashing on lock expirations. What Service Bus consumer tuning should the city apply?
• The library wants bulk re-embed jobs to start overnight without staff starting workers at midnight. How should Service Bus schedule that work?
• When a new ordinance PDF lands in Blob Storage, the city’s RAG indexer should start without polling. Which integration should they use?
• Emergency management requires a human approval signal before publish agents distribute an alert draft. What Event Grid approach fits?
• Only application/pdf blob creations should invoke the city’s OCR-to-embed Function. How should Event Grid be configured?
• A flaky enrichment endpoint drops intermittent Event Grid deliveries for permit metadata. What reliability configuration should the city apply?
• The same ordinance-published event must update analytics logging and invalidate a live agent cache. What Event Grid pattern should the city use?
• Parks chatbot tools need a lightweight internal HTTP API for lookup actions. Which Azure Functions trigger should host that API?
• Staff upload new SOP PDFs that must be chunked for RAG automatically. Which Functions trigger starts that work on upload?
• Code-enforcement summary jobs sit on a Service Bus–backed work queue. How should Azure Functions consume that work?
• When FAQ items change in Cosmos DB, the city’s Redis cache for the resident assistant must refresh. Which Functions approach fits?
• Nightly, the city must find stale embeddings and enqueue re-embed messages. Which Functions trigger schedules that maintenance?
• A permits enrichment Function should write outputs to Blob Storage and enqueue follow-up work with minimal SDK boilerplate. What should developers use?
• The city must stage a new agent-tool Function API before residents hit it. Which deployment practice reduces release risk?
• Long RAG enrichment jobs exceed default Consumption timeouts for the city’s Function app. What hosting decision fits?
• A Function must send to Service Bus without storing a connection string in app settings. What authentication approach should the city use?
• A Function repeatedly fails on a corrupt permit PDF from the queue. What poison-message strategy should operators follow?
• Courts IT creates a new custom Event Grid topic and webhook endpoint for workflow events. What must the endpoint handle during subscription setup?
• A failed agent step must be traced from a Service Bus message through Application Insights. What correlation practice should developers apply?
• A Service Bus topic must route only “public-health” intents to the clinic assistant worker. Which feature implements that rule?
• Event Grid delivers the same blob-upload event twice to the city’s summarizer Function. How should the Function stay idempotent?
• Embedding workers for the library SOP corpus need higher Service Bus throughput without raising lock thrash. What receive pattern helps?
• Summarizer Function deliveries keep landing in Event Grid’s failure destination after repeated errors. What should ops investigate first?
• Permit packet generation needs multi-step generative work with checkpoints and retries across hours. Which Functions approach fits best?
• An internal civic agent calls HTTP-triggered tool Functions. Which auth approach best fits machine-to-machine calls inside the city tenant?
• Thousands of 311 events flood Event Grid during a storm surge into AI triage. What capacity planning concern matters most?
• A Service Bus transaction cannot undo model tokens already spent on a failed civic enrichment. What design lesson applies?
• High-volume water-meter alert paraphrasing needs more Service Bus throughput than a single non-partitioned queue provides. What helps?
• Blob-created events from Azure Storage should start RAG indexing without polling. Which Event Grid topic type is the natural fit?
• An embedding Function app must load connection settings and secrets without hardcoding them in code. What configuration practice fits?
• Clerk desk tools hosted on Azure Functions suffer cold-start delays that frustrate staff. What hosting mitigation fits latency-sensitive tools?
• A legacy city webhook posts nonstandard JSON that AI workflows need on Event Grid. How should Azure Functions help?
• Multiple Function instances must process redaction jobs from one Service Bus queue safely. Which pattern applies?
• After a housing case row commits, an agent follow-up message must not be lost if the process crashes mid-flight. What pattern helps?
• Only blob subjects under /permits/ should wake the AI classifier subscription. How should Event Grid be scoped?
• Sensitive resident-notice drafts must wait for human review before the distribution agent runs. Which Service Bus feature fits?
• A public parks FAQ site calls an Azure Function HTTP API. What exposure practice is appropriate?
• Poison/dead-letter queue depth for document AI jobs is climbing. What operational response fits?
• Parallel enrichment Functions must finish before the city enqueues the final summary job. What coordination pattern fits?
• A simple fire-and-forget park photo caption job needs a basic queue. For AI-200 messaging skills, which service is the primary exam focus?
• After a Function outage, Event Grid retries create a retry storm into the city’s AI handler. What mitigation helps?
• Ephemeral agent-session work queues should disappear after idle timeout. Which Service Bus lifecycle setting helps?
• A grounded-answer Function should persist audit rows to Cosmos DB with minimal custom SDK plumbing. What helps?
• Functions that call internal RAG data services must avoid public internet exposure of those stores. What networking approach fits?
• Agent tool schemas still call /api/v1/tools while a new Function API is ready. How should routes be versioned safely?