Amazon Kendra
Intelligent enterprise search
An ML-powered search service for internal documents: connects to many sources (S3, SharePoint, Confluence…), understands natural-language questions, and returns relevant answer passages; also used as a retriever for RAG.
- Cost$$$ High
- Difficulty●●●●●
- Costs money when idle?Yes
- Managed by AWS
- In examsCLF · SAA
Think of it asLike a librarian who understands questions: ask "how many days of leave do I get" and it points right to the passage in the HR handbook.
When to use it
✓ Use it when
You need a smart search box across many internal document repositories.
✗ Avoid it when
Simple full-text search on a small budget → OpenSearch Service; ready-made Q&A assistant → Amazon Q Business.
Common uses
- Internal policy search
- Retriever for RAG chatbots
- Customer support knowledge portal
What does it cost?
- Pricing model: Charged by index running hours (by edition/capacity unit), plus connector fees for syncing documents.
⚠️ Cost traps
Leaving an index running for test environments is very expensive; delete indexes when not in use.
