使用 AWS Analytics MCP 服务器进行代理数据工程.pdf

1、 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.A N T 3 3 5Agentic data engineering with AWS Analytics MCP ServersRam NottathPrincipal Solutions Architect-Data and AI,AWSHarshida PatelPrincipal Specialist Solution

2、s ArchitectAWS 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.AgendaThe Challenge-Current state of data engineering pain pointsThe Solution-Agentic AI for data engineeringHow It Works-MCP architecture&AWS implementationDemo-Seeing it in actionBest PracticesResources 2025,Amazon W

3、eb Services,Inc.or its affiliates.All rights reserved.2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Meet AnyCompany Retailer 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Mission:data in action-from numbers to ImpactCollectStoreProcessInsightsdata Analytics

4、pipelinedata foundationgrowing use casesgenerative AIcustomer 360data mesh 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Reality check Data engineers sharedHow long to build that new promotion analysis pipeline?“Answer:2-3 weeksWhen can we get real-time inventory insights?Answer

5、:3-5 days behindCan we automate Monday morning reports?Answer:Manual refresh onlyThe Cost:Missed flash sales,inventory issues,slow decisions 2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Pillars of productivity lossD A T A E N G I N E E R SS H A R E DContext SwitchingData Qualit

6、yError Handling&DebuggingResource ManagementSearching for Documentation and best practicesThese challenges were piling up.Then one day,things came to a head.2025,Amazon Web Services,Inc.or its affiliates.All rights reserved.Team,Today was a wake-up call.We missed another Marketing deadline,Operation