1、 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.S M B 3 0 5Maximize ROI:How Octus migrated CreditAI from Azure to Amazon BedrockVishal Saxena(he/him)Chief Technology OfficerOctusVaibhav Sabharwal(he/him)Sr.Solutio
2、ns ArchitectAmazon Web ServicesLavanya Bandari(she/her)Sr.Solutions ArchitectAmazon Web Services 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.AgendaRetrieval Augmented Generation(RAG)OverviewCreditAI migration journeyResultsLearningsAWS migration programs 2025,Amazon Web Servic
3、es,Inc.or its affiliates.All rights reserved.What is Retrieval Augmented Generation?RetrievalAdding the retrieved relevant context to the user prompt,which goes as an input to the foundation modelAugmentationGenerationResponse from the foundation model based on the augmented prompt.Fetches the relev
4、ant content from the external knowledge base or data sources based on a user query 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.RAG use-casesImproved content qualityE.g.,helps in reducing hallucinations and connecting with recent knowledge including enterprise dataContextual ch
5、atbots and question answeringE.g.,enhance chatbot capabilities by integrating with real-time dataPersonalized searchE.g.,searching based on user previous search history and personaReal-time data summarizationE.g.,retrieving and summarizing transactional data from databases,or API calls 2025,Amazon W
6、eb Services,Inc.or its affiliates.All rights reserved.Knowledge bases for Amazon BedrockNATIVE SUPPORT FOR RETRIEVAL AUGMENTED GENERATION(RAG)6MODELAnthropicClaudeMetaLlamaAmazonTitan TextAI21 LabsJurassic2USER QUERYAUGMENTED PROMPTANSWERKNOWLEDGEBASES FORAMAZON BEDROCK321456Securely connect FMs to