A machine learning-powered enterprise search service that enables natural language queries over internal documents (PDFs, wikis, SharePoint), returning precise answers instead of just a list of links.
In AIF-C01 (Domain 3) it is compared to Bedrock Knowledge Bases: Kendra is a standalone semantic search engine, while Knowledge Bases integrates retrieval directly into a foundation model's RAG flow. Both can be used together as a data source.
Amazon Kendra is a concept that can show up on the exam. Check the full guide to see it in context.