《使用 Amazon SageMaker Studio 和 Bedrock AgentCore 更快地开发 AI 代理.pdf》由会员分享,可在线阅读,更多相关《使用 Amazon SageMaker Studio 和 Bedrock AgentCore 更快地开发 AI 代理.pdf(44页珍藏版)》请在三个皮匠报告上搜索。
1、A I M 3 8 8Develop AI agents faster with Amazon SageMaker Studio and AgentCoreSumit Thakurhe/himSr.Manager Product ManagementAmazon Web Services(AWS)Davide Gallitellihe/himSr.WW Specialist SA GenAI/MLAmazon Web Services(AWS)Nikhil Singhalhe/himSr.Staff ML EngineerRobinhood Markets 2025,Amazon Web Se
2、rvices,Inc.or its affiliates.All rights reserved.5 0%DistillationLoRAQLoRA1%Domain-specific modelsCustomization TechniquesAgentic AI adoption is acceleratingReasoning CapabilitiesThinkActObserve 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Agent development workflowGather and p
3、repare data Build,deploy,and monitor agentCustomize and evaluate modelsDeploy modelPlan and set up evaluation criteriaG O V E R N A N C E|R E S P O N S I B L E A I 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Explore our productsAmazon BedrockFast-track Generative AI applicatio
4、ns and agentsGen AI Application DevelopmentAmazon SageMaker AIBuild,train,and deploy AI models for any use caseModel DevelopmentAmazon Bedrock AgentCoreBuild,deploy,and operate agents securely at scaleAgent Development 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Agent developm
5、ent workflowGather and prepare data Build,deploy,and monitor agentCustomize and evaluate modelsDeploy modelPlan and set up evaluation criteriaG O V E R N A N C E|R E S P O N S I B L E A I 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Plan,execute,and reproduce end-to-end model c
6、ustomization workflowAVAILABLE IN PREVIEWNEWSageMaker Model Customization Agent 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.SageMaker model customization agentPlan model customization workflow for your use caseSetup model evaluation criteria,including RAI dimensionsSpec-driven