Search Engine Use Cases - Developer Guide

Desearch can power search products that need fresh web and social context, source links, and AI summaries.

Common use cases

AI answer engines — retrieve current sources, then generate grounded summaries.
Vertical search — search a focused domain, industry, brand, or topic.
Brand and market monitoring — track mentions and trends across web and social sources.
Research workspaces — collect links, posts, and summaries for analysts or agents.
Competitive intelligence — monitor product launches, news, discussions, and technical research.

Recommended endpoints

Need	Endpoint
Multi-source AI search with summary	`POST /desearch/ai/search`
Web links and SERP-style results	`GET /web` or `POST /desearch/ai/search/links/web`
X/Twitter search	`GET /twitter` or `POST /desearch/ai/search/links/twitter`
TikTok, Facebook, Instagram data	Use the relevant endpoint in API Reference.
Raw crawl text	`GET /web/crawl`

Example: AI search request

bash

curl --request POST 'https://api.desearch.ai/desearch/ai/search' \
  --header 'Authorization: $DESEARCH_API_KEY' \
  --header 'Content-Type: application/json' \
  --data '{
    "prompt": "latest AI search API benchmarks",
    "tools": ["web", "reddit", "arxiv"],
    "date_filter": "PAST_MONTH",
    "result_type": "LINKS_WITH_FINAL_SUMMARY",
    "count": 10,
    "streaming": false
  }'

Product guidance

Show source URLs so users can verify answers.
Log request status, latency, and X-Desearch-Cost-Usd for observability.
Handle 401, 422, 429, and 500 explicitly in the UI.
Keep API keys on the server side.