AWS brings RAG evaluation and LLM-as-a-judge feature to Amazon Bedrock

AWS has updated Amazon Bedrock with new features to streamline application testing for enterprises. Key updates include a retrieval augmented generation (RAG) evaluation tool and support for custom connectors and reranking models in Bedrock Knowledge Bases. These enhancements allow for more efficient data ingestion and improved relevance in responses. Additionally, a new LLM-as-a-judge feature helps enterprises perform automated evaluations of AI models, reducing costs and speeding up feedback loops.