1# Azure AI Search — TypeScript SDK Quick Reference
2 
3> Condensed from **azure-search-documents-ts**. Full patterns (semantic config, vector profiles, autocomplete)
4> in the **azure-search-documents-ts** plugin skill if installed.
5 
6## Install
7```bash
8npm install @azure/search-documents @azure/identity
9```
10 
11## Quick Start
12```typescript
13import { SearchClient, SearchIndexClient, SearchIndexerClient } from "@azure/search-documents";
14const searchClient = new SearchClient(endpoint, indexName, credential);
15```
16 
17## Non-Obvious Patterns
18- Vector search uses `vectorSearchOptions.queries` array with `kind: "vector"`
19- Semantic search requires `queryType: "semantic"` + `semanticSearchOptions`
20- Batch ops: `searchClient.indexDocuments({ actions: [{ upload: doc }, { delete: doc }] })`
21 
22## Best Practices
231. Use hybrid search — combine vector + text for best results
242. Enable semantic ranking — improves relevance for natural language queries
253. Batch document uploads — use `uploadDocuments` with arrays, not single docs
264. Use filters for security — implement document-level security with filters
275. Index incrementally — use `mergeOrUploadDocuments` for updates
286. Monitor query performance — use `includeTotalCount: true` sparingly in production
29