Amazon Kendra
An intelligent enterprise search service powered by ML.
Exam Tip: Kendra = intelligent enterprise search. If the question involves searching across enterprise documents with ML-powered relevance, think Kendra. Note: Kendra can also be used as a retrieval component in RAG architectures. Key difference from Q Business: Kendra returns SEARCH results, Q Business provides ANSWERS.
Document Search
- Search across multiple data sources with natural language queries
- Supports incremental learning from user feedback
- Understands context and intent (not just keyword matching)
- Connectors for 40+ data sources (S3, SharePoint, Salesforce, databases, etc.)
- Returns precise answers (not just links to documents)
- FAQ support (automatically extracts Q&A pairs)
Incremental Learning
- Learning from user interactions (click-through data)
- Relevance tuning to improve search results over time
- Manual boosting/burying of specific results
Common Use Cases
- Enterprise knowledge management and document search
- Internal FAQ and help desk systems
- Legal document discovery and research
- Medical literature and research paper search
- Code and technical documentation search