Amazon Kendra
#206 · Level 3 · Modern architecture · AI and machine learningConcept
Section titled “Concept”Amazon Kendra is a smart search inside a company’s own documents. It uses ML to understand the meaning of a question, instead of only matching words. Using connectors, documents from many places such as S3, SharePoint and databases can be indexed in one place. Search follows the user’s permissions, so it shows each person only the documents they have the right to see. Kendra is also used as a source (retriever) that finds the right information for generative AI apps.
Simple example
Section titled “Simple example”An office librarian who listens to your question and shows you exactly which part of which document has the answer.
Why it’s needed
Section titled “Why it’s needed”Documents are scattered in many places; normal search matches words but does not understand meaning.
Where it’s used
Section titled “Where it’s used”Employees looking up policies, support knowledge bases, giving AI chatbots the right information.
Azure equivalent
Section titled “Azure equivalent”Azure AI Search
Learn more
Section titled “Learn more”For details and the latest information, see the official documentation: