Implement AI capabilities in database solutions

DP-800 · 50 questions

• A city evaluates external embedding models for multilingual 311 notes and must weigh model size against structured-output needs. What should the team do first?
• A public-health clinic needs models that can describe photo attachments from field visits, not only plain-text case notes. Which model choice fits?
• A permits bureau wants Azure SQL to call a registered embedding model for fee-policy text. What must developers create and manage?
• A housing authority needs embeddings refreshed when case narrative columns change without putting heavy work on every OLTP write. Which maintenance approach should they choose?
• A 311 desk needs embeddings refreshed immediately after request description edits. Which maintenance pattern fits that requirement?
• Transit ops want route-alert embeddings regenerated asynchronously when alert rows change in SQL. Which Azure pattern should they use?
• Courts IT wants nightly embedding rebuilds only for opinions that actually changed. How should they combine change capture with orchestration?
• Emergency management needs incident text changes fan out to an embedding worker without polling the database. What should they prefer?
• An open-data team is choosing a managed path to keep published dataset blurb embeddings current. Which option should they evaluate for that maintenance workflow?
• A library plans semantic search over the catalog and must pick which columns feed embeddings. What should they identify as inputs?
• A water utility must embed long lab narrative PDFs without exceeding model token limits. What design step is required?
• After the county clerk configures an external model, statute summary columns still lack vectors. What should they do next?
• Public-health plans to embed case narratives but must protect sensitive identifiers. Which guidance is correct?
• Parks policy documents have overlapping sections that must not lose context at chunk boundaries. What chunking strategy should they use?
• A housing authority wants amenity tags extracted as structured fields before those tags are embedded. What model capability should they evaluate?
• 311 receives English and Spanish resident notes and needs embeddings that respect language differences. How should external models be managed?
• Transit OLTP peaks make synchronous embedding work on every write too costly. Which maintenance tradeoff should they choose?
• County SQL AI developers must register an external model so T-SQL can call it. What configuration is required?
• Court clerks need exact statute section keyword lookups such as a precise section number string. Which search approach should they choose?
• 311 wants classic keyword search over request summaries in SQL. What should developers implement?
• Housing designers must store case-note embeddings in SQL for semantic search. What should they design for the vector columns?
• A library needs faster similarity search over catalog blurb embeddings. What should they create?
• Permits analysts want to rank fee-policy chunks by closeness to a natural-language question embedding. Which T-SQL function supports that ranking?
• A water utility compares lab-narrative embeddings and must avoid distance skew from unnormalized vectors. What should they apply before comparisons?
• Emergency management runbooks need to verify vector and index metadata in the ops database. Which function should they use?
• An open-data portal needs nearest-neighbor retrieval over dataset description embeddings in SQL. Which function should developers call?
• Transit must search a very large ridership-comment embedding corpus where approximate neighbors are acceptable. Which approach should they choose?
• A courts clerk must evaluate vector index types and distance metrics before opinion semantic search goes live. What should the team do?
• Public-health staff need semantic lookup over clinic FAQ embeddings inside SQL. What should developers implement?
• City 311 wants retrieval that combines keyword full-text hits with vector similarity over knowledge articles. What should they implement?
• Housing merges keyword and vector ranked lists for lease-policy assistants. Which fusion technique should they apply?
• Parks compares vector-only and hybrid search on recreation-policy documents before choosing a production pattern. What should they evaluate?
• Courts need exact nearest neighbors for a small docket-note embedding set where approximate results are not acceptable. Which search mode should they select?
• A library needs synonym-aware keyword search so “storytime” matches children’s program catalog names. What should they configure?
• A water utility needs hybrid search that still elevates exact chemical-name keyword hits while using vectors for narrative lab notes. How should they design it?
• Emergency management added a new vector index type and must check VECTORSEARCH responsiveness. What should operators measure?
• Permits store embedding vectors beside chunk text and must keep dimensions compatible with the chosen model. What should designers document and enforce?
• County residents ask fuzzy questions such as “why was my street flooded” where exact keyword matches are weak. Which search approach should the team favor?
• City 311 wants assistants that answer ordinance questions grounded in retrieved city policy text rather than free-form invention. Which pattern fits?
• Housing must call a language model from SQL as part of a RAG prompt flow for lease FAQs. Which procedure supports sending those prompts?
• Courts retrieve structured docket rows that must be packaged for language-model context in a RAG flow. What should developers do first?
• Public-health RAG must give the model both retrieved clinic-policy chunks and the staff question. What should the solution send?
• Transit receives a language-model payload for the rider-alert assistant and must surface the natural-language answer. What should developers do?
• A library RAG assistant must reduce hallucination risk on catalog-policy answers. What grounding discipline should they enforce?
• Water ops want sensor anomaly rows included in a RAG prompt’s context window for the language model. How should those rows be packaged?
• Emergency shelter SOPs change often and the team needs fast policy updates without model retraining. Which approach should they choose?
• Permits call an external language-model REST endpoint from SQL and must authenticate securely. What should they use?
• 311 receives structured JSON from the language model that includes reply text and citation chunk IDs. What should the app do?
• Housing residents need live transactional fee quotes calculated by existing stored procedures, not narrative policy summaries. When should the team avoid RAG?
• A courts clerk must summarize several related ordinance chunks with a language model in one RAG call. What should they assemble first?