Agentic retrieval with LangChain and Amazon Bedrock Knowledge Bases
Amazon Bedrock now offers agentic retrieval that breaks multi-part queries into sub-queries automatically
AWS introduced agentic retrieval for Amazon Bedrock Managed Knowledge Bases, which automatically decomposes complex multi-part questions into sub-queries, executes them iteratively, and verifies coverage before returning results. This addresses a known weakness in standard RAG where single-vector similarity search fails to capture all intents in compound questions. The capability is exposed via a new AgenticRetrieveStream API and integrated into the langchain-aws package.