Know How Guide and Hands on Guide for AWS
개요
- Summary of the Amazon DynamoDB Service Disruption in Northern Virginia (US-EAST-1) Region - Building hexagonal architectures on AWS - Service Screener is a tool that runs automated checks on AWS environments and provides recommendations based on AWS and community best practices 本人与 Kiro Crew agent 一起动手完成、验证过的端到端实验(均含可复现资产 / IaC / 实测记录): - — AWS Lambda MicroVMs(Firecracker 隔离的 serverless,2026-06 GA)动手实验。单 Flask MicroVM 生命周期(create-image → run → connect → suspend/resume,验证快照状态存活)+ 多租户 aws-sample 端到端部署;实测记录、连接契约(443 + X-aws-proxy-auth + X-aws-proxy-port)、跨代状态丢失的 mount-namespace 根因分析,findings 已回报 aws-samples 上游 issue。 - — DuckDB 7 个场景(NYC 出租车分析、httpfs/S3 远程 Parquet、CSV→Parquet ETL、NL→DuckDB Chat BI、Pandas 零拷贝、日志分析、多 agent AgentCore 设计),spec 驱动 + 34 用例 pytest + CI;场景 04+07 真实部署到 Bedrock AgentCore Runtime(embedded DuckDB 读 S3 hive-partitioned Parquet)验证后清零。
README
aws-is-how
- aws-is-how
- 常见故障排除及支持手册
- AWS Skill builder
- freeCodeCamp
- Architecture Design
- Hands-on Labs (自建实验)
- Introducing the AWS Product Lifecycle page and AWS service availability updates
- AI/ML
- ML Study
- SageMaker
- Jupyter Notebooks
- Compute vision
- GenAI Overview
- Promote-Engineering
- Video in GenAI
- Audio in GenAI
- Image in GenAI
- Scenario based GenAI
- Vector database
- Agent database
- LLM Models
- World Models
- Bedrock practice
- RLHF
- Knowledge Base
- Deep Research
- AI Coding
- Agent workflow and Framework
- MCP
- Agent Infra
- Agent Technology
- OpenClaw
- GenAI Business Application
- Federated ML
- ML Hardware
- Robotics
- Cost
- Computing
- Analytics
- High Level Data Engineering and Data Analytics
- Data integration service: Glue
- Analysis: EMR
- Data On EKS
- Stream - Flink and Spark Streaming
- Stream - Kinesis
- Stream - Kafka
- Ad-hoc and Interactive query: Athena
- Data Warehouse: Redshfit
- Search and analytics: Elasticsearch Service
- Governance
- BI
- Delta Lake
- IOT
- Security
- Encryption - KMS
- Credential - Secret Manager
- Certificate - Certificate Manager
- Asset Management and Compliance
- AuthN and AuthZ
- Sentitive Data
- Threat detection - GuardDuty
- WAF
- Permission - IAM Policy, S3 Policy, RAM Policy
- Multi accounts structure
- SIEM and SOC
- Vulnerability Assessment - Inspector and Alternative
- Network
- DNS
- Serverless
- Migration
- Modernization
- Storage
- Database
- Container
- DevOps
- Integration
- Media
- Mobile
- Business continuity
- Game
- SAP
- Office and business application
- Metaverse
- Automotive
- HealthCare and Life Science
- Retail and ECommerce
- FSI
- Hospitality
常见故障排除及支持手册
AWS Skill builder
freeCodeCamp
Architecture Design
- Building hexagonal architectures on AWS
- Service Screener is a tool that runs automated checks on AWS environments and provides recommendations based on AWS and community best practices
Hands-on Labs (自建实验)
本人与 Kiro Crew agent 一起动手完成、验证过的端到端实验(均含可复现资产 / IaC / 实测记录):
- Lambda MicroVMs Lab — AWS Lambda MicroVMs(Firecracker 隔离的 serverless,2026-06 GA)动手实验。单 Flask MicroVM 生命周期(create-image → run → connect → suspend/resume,验证快照状态存活)+ 多租户 aws-sample 端到端部署;实测记录、连接契约(443 +
X-aws-proxy-auth+X-aws-proxy-port)、跨代状态丢失的 mount-namespace 根因分析,findings 已回报 aws-samples 上游 issue。 - DuckDB Scenarios Explorer — DuckDB 7 个场景(NYC 出租车分析、httpfs/S3 远程 Parquet、CSV→Parquet ETL、NL→DuckDB Chat BI、Pandas 零拷贝、日志分析、多 agent AgentCore 设计),spec 驱动 + 34 用例 pytest + CI;场景 04+07 真实部署到 Bedrock AgentCore Runtime(embedded DuckDB 读 S3 hive-partitioned Parquet)验证后清零。
- COD-Commerce Agent — COD(Cash on Delivery / 货到付款)单页电商的对话式购物 agent:Bedrock AgentCore Runtime 编排(Luna→Claude→规则链 + 工具)、三层记忆(短期事件 / Aurora 确认偏好 / AgentCore USER_PREFERENCE 推断层)、Aurora Serverless v2 + pgvector、S3+CloudFront 单页前端;含 VPC/Aurora terraform、E2E 测试 + CI 与完整实验日志。
- StudyLens on ECS Fargate + EFS + ALB + CloudFront — 把有状态 npm 应用部署为免运维、低成本、手机/iPad 可访问的服务;容器内 adapter 将 OpenAI 兼容调用转成 Bedrock Converse(
us.openai.gpt-5.6-luna,凭 task role 免密钥),CloudFront 提供免域名可信 HTTPS。含完整 Terraform IaC、架构说明与部署/运维/拆除文档。 - AWS Agent Registry Hands-on Lab — AWS Agent Registry(Amazon Bedrock AgentCore 能力,2026-08-31 GA)动手实验:对真实 API 跑通完整生命周期(create → publish → submit → approve/reject → 语义搜索 → teardown 清零)。含 6 个编号脚本 + 一键
end_to_end.py+ 自包含可视化 SPA(Publisher/Curator/Consumer 三列 + 状态机 + 治理前后对比)。实测记录并纠正了两个 API 契约坑:治理面用 clientagent-registry-control,发现面必须用 GA 的agent-registryclient +search_discoverable_registry_records(旧 preview 的bedrock-agentcorenamespace 会对 GA registry 报 404);UpdateRegistryRecordStatus强制要求statusReason。被 Curator 拒绝的记录对 Consumer 搜索不可见——治理价值的核心演示。
Introducing the AWS Product Lifecycle page and AWS service availability updates
AI/ML
ML Study
SageMaker
- SageMaker-Workshop
- SageMaker Learning Series
- [SageMaker Notebook]
- [SageMaker training job and HyperPod]
- SageMaker input mode: pipe mode and file mode
- Save costs by automatically shutting down idle resources within Amazon SageMaker Studio
- SageMaker Neo supported devices edge devices
- SageMaker-HyperPod-InstantStart
- Amazon SageMaker HyperPod introduces Amazon EKS support
- Amazon EKS support in Amazon SageMaker HyperPod
- Hyperpod Troubleshooting Guide
Jupyter Notebooks
Compute vision
- Use SageMaker for Automotive Image Classification
- ML Bot Workshop
- IP Camera AI SaaS Solution
- image classification using resnet
- Open CV on Lambda
- [OCR]
- 新希望-构建云上智慧牧场
GenAI Overview
- 一文读懂AIGC
- 跨模态深度学习模型CLIP(Contrastive Language-Image Pre-Training)
- “对抗生成网络”GAN(Generative Adverserial Network)
- Diffusion模型
- Token, Embeding, Self-Attention, Transformer, Vector, Encoding output 101
- ChatGPT Overview
- AIGC workshop
- 三类场景赋能行业创新
- A guide to making your AI vision a reality
- understand The tokenize
- 图解AI三大核心技术:RAG、大模型、智能体
- 红杉 AI Ascent 2026 主题演讲全文译, 2026红杉美国年度分享:AGI已经到来,重新定义AI革命
Promote-Engineering
- Prompt engineering techniques and best practices with Claude3
- Implementing advanced prompt engineering with Amazon Bedrock
- Evaluating prompts at scale with Prompt Management and Prompt Flows for Amazon Bedrock
- Enhance performance of generative language models with self-consistency prompting
- Context Engineering 上下文工程
Video in GenAI
- [Video Understanding and Management]
- [Generate Video]
Audio in GenAI
- 构建文生音场景定制化人声解决方案
- 构建实时音视频交互解决方案-TEN-Agent and Nova
- 基于Amazon Bedrock 构建端到端实时语音助手 and Amazon Bedrock的实时语音解决方案
- new-alexa-generative-AI
- MiniMax Speech 2.5 3秒Clone你的声音,40种语言完美还原
- [Content Moderation]
Image in GenAI
- [Stable Diffusion]
- stable-diffusion-webui self hosted on g4dn.xlarge with Ubuntu 22.04 LTS
- Remember run
sudo apt-get updatebeforesudo apt install wget git python3 python3-venv. - Run
ssh -L 7862:localhost:7862 [email protected]orbash stable-diffusion-webui/webui.sh --share - Install Nvida Cuda
- create-your-own-stable-diffusion-ui-on-aws
- Remember run
- sagemaker-stablediffusion-quick-kit
- Stabule Diffusion on EKS
- Stable Diffusion Quick Kit 动手实践 – 基础篇
- Stable Diffusion Extention hosting on AWS
- SageMaker Notebook 机器学习服务轻松托管 Stable Diffusion WebUI
- Stable Diffusion on Amazon SageMaker Workshop
- inpaint-images-with-stable-diffusion-using-amazon-sagemaker-jumpstart
- stable-diffusion-webui self hosted on g4dn.xlarge with Ubuntu 22.04 LTS
- open_artifacts_for_bedrock
- SageMaker LMI+Streaming 构建 端到端GenAI Text2Image应用
- [ComfyUI]
- [Deployment]
- [Integration and Usage]
- [Use case]
- Amazon Nova 生成图片 Prompt
- [GPT-Image-2]
Scenario based GenAI
- [Chatbot]
- [Text2SQL - Data Agent]
- DB-GPT与百川社区强强联手,无缝支持百川模型推理与Text2SQL微调
- text2sql_gen_demo notebook
- Generative BI using RAG on AWS
- Data engineering agent - 面向业务的 AI Chatbot、面向系统的 Agent API,以及面向分析的 BI Dashboard,统一的 Agent 被交付
- 云器 Data Agent
- AI Agent for Data - Data Agent
- Dagster is a data orchestrator built for data engineers, with integrated lineage, observability, a declarative programming model, and best-in-class testability
- [Digital human and Role play]
- [OCR]
- [VOC]
- [Content moderation]
- [Translate]
- [Traditional NLP]
- [Forecasting]
- [Fraud Detection]
- [Recommandation]
- [Prediction Maintenance]
- [Contact Center]
- [Labeling]
- [CV Labeling]
Vector database
- RDS for PostgreSQL now supports pgvector for simplified ML model integration
- FlagEmbedding - retrieval, classification, clustering, or semantic search. And it also can be used in vector databases for LLMs
- S3 Vectors: First cloud storage with native vector support at scale
Agent database
- [DuckDB]
LLM Models
Claude
- [Claude Workshop]
- Amazon Bedrock Claude3 Workshop
- Claude Artifacts workshop
- Fine-tune Anthropic’s Claude 3 Haiku in Amazon Bedrock to boost model accuracy and quality
- Anthropic API Gateway Proxy: 无需修改代码即可在 Anthropic SDK 中使用 AWS Bedrock 上的各种大语言模型
- Opus 4.7 + claude code 问题汇总 2026.04.20
- From developer desks to the whole organization: Running Claude Cowork in Amazon Bedrock
- Claude Platform on AWS: Anthropic’s native platform, through your AWS account
- 斯坦福 STORM 方法:怎样让 Claude 在几分钟内像博士一样做研究
OpenAI GPT
- [GPT-OSS]
- [GPT 闭源模型]
Llama
Nova
- Amazon Nova Multimodal understanding workshop
- 用 Amazon Bedrock 与 Nova 大模型构建客户之声解决方案
- Amazon Nova Canvas and Amazon Nova Reel Workshop
- Nova Sonic Speech-to-Speech Model Samples
- Nova MME nova-multimodal-embedding-model practice
- Amazon SageMaker Inference for custom Amazon Nova models
- [和 Nova Forge 是否有结合,或者两者如何进行定位? 这里提的是指轻量的微调,不是那种有数据混合的, 数据混合的还是采用 Nova Forge]
Deepseek
- Deepseek on AWS deployment
- deploy-deepseek-r1-distilled-llama-models-with-amazon-bedrock-custom-model-import
- 轻松部署DeepSeek-R1 671B动态量化模型
- Graviton4 Run Deepseek
- Build agentic AI solutions with DeepSeek-R1, CrewAI, and Amazon SageMaker AI
- 使用亚马逊云科技自研芯片 Inferentia2 部署 DeepSeek R1 Distillation 模型(一)
- 使用亚马逊云科技自研芯片 Inferentia2 部署 DeepSeek R1 Distillation 模型(二)
- 本地 671B DeepSeek-Coder-V3/R1: 仅使用 14GB 显存和 382GB 内存运行其 Q4_K_M 版本
- Hosting DeepSeek-R1 on Amazon EKS
China otherLLM
More Open-Source Model
- ColossalAI for LLM quick training
- [Gemma]
- [Grok]
- [Kimi]
- Kimi K3 on Amazon Bedrock Example
- 用 Kimi K3 做 Coding Assistant: 单文件登月游戏实战 - OpenCode/OpenAI 兼容端点接入、提示词与迭代过程、Converse API 坑、无头验证脚本
- moon_lander.html 游戏本体 - 单文件无依赖,方向键控制姿态与升降,需在崎岖月面选择平坦着陆点并保持箭体直立
vllm and ollma
- vllm_quickstart
- lightweight tool designed to simplify models deployment
- Testing LLM API performance with automated deployment and extensive test matrices: Throughput, First Token Latency, Percentiles, End-to-End Latency, Output Tokens Per Second
World Models
Bedrock practice
- Patterns for Building Generative AI Applications on Amazon Bedrock
- 使用 Amazon SageMaker 和 Bedrock 构建营销场景端到端应用
- 使用 Amazon Bedrock 和 Amazon SageMaker,开启全新的生成式 AI
- 构建端到端生成式 AI 应用
- Amazon Bedrock Workshop
- Amazon Bedrock Development Workshop - GCR
- open_artifacts_for_bedrock
- GenAI Serverless ImmersionDay
- streaming response from Amazon Bedrock with FastAPI on AWS Lambda
- 为生成式 AI 产品打造持续的卓越用户体验——跨区域高可用弹性解决方案
- Why Claude 4 API Hits Rate Limits: Token Burndown Explained
- Use AWS PrivateLink to set up private access to Amazon Bedrock
- 基于 Amazon SageMaker 和 LLaMA-Factory 打造一站式无代码模型微调部署平台 Model Hub
- Bedrock API Key
- Bedrock 多模型接入 – Bedrock Connector 部署与使用指南
- [Prompt Caching]
- 使用Amazon Bedrock + 自建ECS Docker Sandbox实现Agent 程序化工具调用Programmatic Tool Calling
RLHF
- Align Meta Llama 3 to human preferences with DPO
- Thinking-Claude make Claude as GPTo1
- Residual Connections Are Getting an Upgrade (mHC)
Knowledge Base
RAG - retrieval-augmented generation
- Smart Search 基于智能搜索的大语言模型增强方案
- Smart Search V2 基于智能搜索的大语言模型增强方案2
- 基于智能搜索和大模型打造企业下一代知识库
- 基于智能搜索和大模型打造企业下一代知识库 之 制造/金融/教育/医疗行业实战场景
- 基于 RDS 和 Confluence 数据源构建端到端的RAG
- 基于大语言模型和推荐系统构建电商智能导购机器人
- 基于大语言模型知识问答应用落地实践 – 知识召回调优
- 基于LLM 和 Amazon Opensearch 或 Amazon Kendra 打造企业私有知识库
- 基于Amazon Open Search+大语言模型的智能问答系统
- GenAI Data Foundation Workshop - Healthcare RAG chatbot
- RAGChecker for RAG health check
- New APIs in Amazon Bedrock to enhance RAG applications
- RAG 挑战赛冠军方案解析:从数据解析到多路由器检索的工程实践
- Build a RAG based question answer solution using Amazon Bedrock Knowledge Base, vector engine for Amazon OpenSearch Service Serverless and LangChain
- Create an agentic RAG application with LlamaIndex, and Mistral in Amazon Bedrock
LLM based KB
- karpathy A pattern for building personal knowledge bases using LLMs - LLM Wiki
- WeKnora — 让文档活起来:RAG、Agent 推理与自动 Wiki 一体化的知识框架
Deep Research
AI Coding
- [AIDLC-AIPLC-AgentDLC]
- [Ontoplogy on AI Coding]
- [Refactor existed code]
- [Amazon Q Developer/Kiro]
- Kiro Community
- Amazon Q & CodeWhisperer for VS Code
- Q Developer Workshop
- Agentic AI 帮你做应用 —— 从0到1打造自己的智能番茄钟
- Agentic AI 帮你做应用 —— BotGroup 吵架机器人
- Amazon Q CLI + MCP 创建 AWS 架构图
- Prompt Driven Development Workshop
- Proxy gateway for Kiro API
- 配置 kiro 使用企业单点登录系统
- 通过ACP协议,使用api调用kiro cli - 把 Kiro CLI 当作 Agent SDK:一键订阅即可构建你的Agent应用
- Benchmarking Kiro CLI coding-agent performance and cost across multiple models, replicating the Artificial Analysis Coding Agent Index methodology
- [Claude Code]
- Claude3 code assistant
- Claude Code on Amazon Bedrock: Quick Setup Guide
- autoresearch 方法,把 skill 通过率从 50% 提到了 90%
- Claude Code Demo for LS_HC Enterprise customer
- Everything claude code
- 我给 Claude Code 做了个 AI 硬件监工 - claude-desktop-buddy M5 Paper Buddy
- Claude Code Security Reviewer
- cc-connect Bridge AI Agents to Chat Platform with ACP
- Remote Claude Control by Feishu and Tmux
- A field guide to Claude Fable 5: Finding your unknowns
- [GenDev for SRE - AIOPs Agent or SRE Agent]
- [SDE Agent and DevOps Agent]
- MultiAgent - ChatDev on Claude 3 一句话实现一个软件需求
- SWE-agent turns LMs (e.g. GPT-4) into software engineering agents
- 在 EKS 上部署 Outline Wiki,集成完整的可观测性方案(Prometheus + Grafana + OpenSearch)以及 AWS DevOps Agent。包含飞书通知和故障注入脚本,用于现场演示场景
- DevOps Agent exposed as ACP and MCP
- 把 AWS DevOps Agent 连接到部署在 EKS 上的自建 MCP Server,通过 VPC Lattice Private Connection 在私网完成 AWS 中国区以及其他云的管理
- 将阿里云 云效(Yunxiao) 官方 MCP server,以容器形式托管到 Amazon Bedrock AgentCore Runtime,对外暴露为一个受 IAM/SigV4 保护的远程 MCP 端点,供 AWS DevOps Agent 等 MCP 客户端调用
- AWS DevOps Agent adds release management capabilities
- Automated network incident response with AWS DevOps Agent with SNS and Lambda
Agent workflow and Framework
- AWS Multi-Agent-Orchestrator - Agent Squard - Flexible and powerful framework for managing multiple AI
- AWS announces new innovations for building AI agents at AWS Summit New York 2025
- [LangChain and LangGraph]
- Dify - an open-source large language model (LLM) application development platform
- [Other Agent tools]
- [Amazon Bedrock Agents]
- Strands Agents
- Strands Agent Hands On
- Open Protocols for Agent Interoperability Part 1: Inter-Agent Communication on MCP
- Learning-Strands-Agents
- Introducing Strands Agents, an Open Source AI Agents SDK
- 亚马逊云科技中国区构建 Agentic AI 应用实践指南
- Strands Agent Example
- agenticai-memory-patterns-with-strands-sdk
- Strands Agents SDK 1.0
- Agentic AI on EKS Workshop
- Strands Agents Open Studio
MCP
- [MCP official site](https://github.com/modelcontextprotocol/servers)
- [MCP Servers Explained: What They Are, How They Work, and Why Cline is Revolutionizing AI Tools](https://cline.bot/blog/mcp-servers-explained-what-they-are-how-they-work-and-why-cline-is-revolutionizing-ai-tools)
- [Anthropic 发布了 Streamable HTTP](https://mp.weixin.qq.com/s/9y-VBbP31I8wXur5vEN4Ug)
- [使用 Amazon Lambda 快速部署 Streamable HTTP Github MCP Server](https://aws.amazon.com/cn/blogs/china/deploy-streamable-http-github-mcp-server-using-amazon-lambda/ l)
- [MCP Server hosting]
- [GCR MCP on Serverless](https://github.com/aws-samples/sample-serverless-mcp-server)
- [GCR Sample MCP Servers](https://github.com/aws-samples/aws-mcp-servers-samples)
- [AWS MCP Server List](https://github.com/awslabs/mcp/)
- [Introducing AWS Serverless MCP Server blog](https://aws.amazon.com/cn/blogs/compute/introducing-aws-serverless-mcp-server-ai-powered-development-for-modern-applications/)
- [Playwright MCP Server on Fargate](https://www.notion.so/Playwright-MCP-Server-AWS-Fargate-EKS-1de408b04809808c8604f56fef3cf565?pvs=4)
- [Automating AI-assisted container deployments with the Amazon ECS MCP Server](https://aws.amazon.com/cn/blogs/containers/automating-ai-assisted-container-deployments-with-amazon-ecs-mcp-server/)
- [Accelerating application development with the Amazon EKS MCP server](https://aws.amazon.com/blogs/containers/accelerating-application-development-with-the-amazon-eks-model-context-protocol-server/)
- [MCP Demo]
- [MCP on Amazon Bedrock](ai-ml/chatgpt/claude/mcp/demo_mcp_on_amazon_bedrock.md)
- [sample-agentic-ai-web](https://github.com/aws-samples/sample-agentic-ai-web)
- [A sample MCP server for understanding cloud spend](https://github.com/aws-samples/sample-cloud-spend-mcp-server)
- [Use Case]
- [从零构建 MCP 架构下的 Agentic RAG 系统](https://mp.weixin.qq.com/s/11jlAQkL008Tuq_vEyjjTg)
- [飞书 Remote MCP 配置](https://amzn-chn.feishu.cn/wiki/B5aawkRUdiusxgkvtkNcXAfDnkd)
Agent Infra
- [E2B]
- [Kata Container]
- FlexAI - Architect and deploy advanced Agentic AI platforms on Amazon EKS
- [Lambda MicroVM]
- Lambda MicroVMs 适用场景与 Samples 总结
- AWS Lambda MicroVMs core concepts (https://aws.amazon.com/blogs/aws/run-isolated-sandboxes-with-full-lifecycle-control-aws-lambda-introduces-microvms/)
- Claude Self-Hosted Sandboxes on AWS Lambda MicroVMs
- [sample-multi-tenant-ai-agents-on-lambda-microvm] (https://github.com/aws-samples/sample-multi-tenant-ai-agents-on-lambda-microvm)
- sample-multi-tenant-openclaw-on-firecracker
- [Firecracker]
- [Bedrock AgentCore]
- introducing-amazon-bedrock-agentcore
- AgentCore use case examples
- AgentCore Launch Pad
- New in Amazon Bedrock AgentCore: Build agents with broader knowledge and continuous learning
- [Agent Core Hands on Lab]
- QuickStart
- Above quick start based on sample-strands-in-5-minutes bedrock-agentcore-integration
- bedrock-agentcore-samples
- Deep Dive
- Above Deep Dive based on Diving Deep into Bedrock AgentCore
- QuickStart
- [Migrate to AgentCore]
- Introducing Amazon Bedrock AgentCore Identity: Securing agentic AI at scale
- Introducing Amazon Bedrock AgentCore Gateway: Transforming enterprise AI agent tool development
- Amazon Bedrock AgentCore Memory: Building context-aware agents
- Amazon Bedrock AgentCore Policy
- Amazon Bedrock AgentCore Evaluation & Optimization
- AgentCore introduces new optimization capabilities to continuously improve agents in production
- Eval-First: 基于 AgentCore 构建企业级 Agent (https://github.com/aws-samples/sample-eval-first-building-enterprise-agents-with-agentcore/blob/main/README.zh-CN.md)
- 企业智能体之旅:为什么评估(Evaluation)是一切的起点
- 企业智能体之旅: 评估企业级智能体:从原型验证到生产就绪
- 企业智能体之旅: 如何在亚马逊云科技上构建企业级智能体
- Agent 优化端到端全流程(f7TmsWkmyqkjp5Brtt5V)
- AgentCore Harness - Go from idea to production-grade agent in minutes
- [Agentic AI基础设施实践经验系列]
- Agentic AI基础设施实践经验系列(一):Agent应用开发与落地实践思考
- Agentic AI基础设施实践经验系列(二):专用沙盒环境的必要性与实践方案
- Agentic AI基础设施实践经验系列(三):Agent记忆模块的最佳实践
- Agentic AI基础设施实践经验系列(四):MCP服务器从本地到云端的部署演进
- Agentic AI基础设施实践经验系列(五):Agent应用系统中的身份认证与授权管理
- Agentic AI基础设施实践经验系列(六):Agent质量评估
- Agentic AI基础设施实践经验系列(七):可观测性在Agent应用的挑战与实践
- Agentic AI基础设施实践经验系列(八):Agent应用的隐私和安全
- Agentic AI基础设施实践经验系列(九):Context Engineering 上下文工程
- [AgentCore in Practice]
- [Agent Registry]
Agent Technology
- [Agent to Agent - A2A]
- [Computer use and brower use]
- [Agent Skill]
- [Agent Client Protocol (ACP)]
- [Agent Harness Engineering]
OpenClaw
- [Tech Parts]
- [Business Part]
- [Digital Employee]
GenAI Business Application
- [Quick Suite]
- [Quick Desktop]
- quick-dingtalk-mcp
- feishu-quick-mcp
- 用 Amazon Quick Desktop 实现小红书从选题到发布的全流程自动化
- Amazon Quick Web & Desktop 企业级 Agentic AI 实战工作坊
- [Quick 企业版如何配置desktop 和web 使用同一个账号登陆]
- Amazon Quick 身份体系选型与 SSO 集成完整指南
- Amazon Quick IDP 集成概览
- 如何利用IAM Identity Center用户直接登录Quick Desktop - 主要使用 IAM Identity Center + Amazon Cognito
- 使用 Keycloak 作为 IDP 实现 Amazon Quick SSO 配置指南 - 对于没法使用Amazon Cognito
- Okta + IAM Identity Center 配置指南
- Quick 开通订阅与 Google Workspace SSO 完整手册
- 钉钉 IDaaS 对接 Amazon Quick Desktop OIDC 配置方案
- Amazon_Quick_Desktop_飞书OIDC对接方案
- Entra ID + IAM Identity Center 配置指南
- [Frontier Agent]
Federated ML
ML Hardware
- Hands-on Deep Learning Inference with Amazon EC2 Inf1 Instance
- NVIDIA Dynamo is a high-throughput low-latency inference framework
Robotics
- Genesis-Embodied-AI
- NVIDIA Isaac Lab on AWS
- 云端机器人研发:在 AWS 实现 ROS 2 设备与 Isaac Sim 的 Lerobot 仿真及数据流
- Accelerate Physical AI Development from Prototype to Production with Amazon Kiro
Cost
Cost Explorer
Network cost
Tagging
-
[Tagging when instance and object created]
Sustainablity
Computing
EC2
-
[Linux Daily usage]
-
[Performance]
-
[Windows Daily usage]
-
[GPU Daily usage]
- GPU Onboarding 知识中枢
- How to build Graphics Workstation on Amazon EC2 G4 Instances
- Deploying Unreal Engine Pixel Streaming Server on EC2
- GPU Perf - LLM GPU推理性能可视化平台
- Chip and Morta - AWS Chip Strategy
- [GPU 虚拟化]
- [推理加速 Inference Accleration]
-
[Network of EC2]
-
[Graviton]
- Graviton 2 workshop (https://github.com/aws-samples/graviton2-workshop)
- AWS Workshop - Graviton2 China
- porting-advisor-for-graviton
- AWS Graviton2-based services
- 3rd party
- container
- GRAVITON2 电商独立站
- Porting Advisor for Graviton
- AWS Graviton4 加速 Apache APISIX 消息处理
- Q CLI for Aurora Graviton4 upgrade
- Graviton 官方有个 Kiro Power 做兼容性扫描和出报告
-
[Operation]
- What does :-1 mean in python
- Change EC2 Time-Zone
- How can I set up a CloudWatch alarm to automatically recover my EC2 instance?
- Move EC2 instance to other AZ
- Best practices for handling EC2 Spot Instance interruptions
- How to share the EC2 AMI
- Copy AMI from global to China
- How to handle EC2 detected degradation
- Check if a reboot is required after installing Linux updates
Load Balancer
System Manager
aws ssm get-parameters-by-path --path /aws/service/global-infrastructure/regions --output json --profile us-east-1 --region us-east-1 | jq '.Parameters[].Name'
aws ssm get-parameters-by-path --path /aws/service/global-infrastructure/regions --output json --profile us-east-1 --region us-east-1 | jq '.Parameters[].Name' | wc -l
HPC
-
Orchestrating high performance computing with AWS Step Functions and AWS Batch
-
[NICE DCV]
- NICE DCV Guide
- NICE DCV Connection Gateway - enables users to access a fleet of NICE DCV servers through a single access point to a LAN or VPC
- NICE DCV Session Manager - the Agents, a Broker and API that makes it easy to build front-end applications that programmatically create and manage the lifecycle of NICE DCV sessions across a fleet of NICE DCV servers
- amazon-dcv 2024 launched
-
SOCA
Analytics
High Level Data Engineering and Data Analytics
-
Modern Data Lakehouse Medallion Architecture on AWS with Iceberg Tables
-
Harness the power of your data with AWS Analytics with Lake House
-
Volkswagen streamlined access to data across multiple data lakes using Amazon DataZone
Data integration service: Glue
-
[ETL]
- Quick demo for Glue ETL + Athena + Superset BI
- Glue ETL for kinesis / Kafka and RDS MySQL
- Update and Insert (upsert) Data from AWS Glue
- Introducing PII data identification and handling using AWS Glue DataBrew
- Best practices to scale Apache Spark jobs and partition data with AWS Glue
- monitor optimize cost glue spark
-
[Glue Crawler]
-
[Workflow]
-
[Catalogs]
-
[Delta Lake]
Analysis: EMR
-
[Develop Code]
- EMR Notebooks and SageMaker Use EMR notebooks to prepare data for machine learning and call SageMaker from the notebook to train and deploy a machine learning model.
- Submit EMR Job remotely
- 如何优雅地提交一个 Amazon EMR Serverless 作业?
- 如何在 Amazon EMR Serverless 上执行纯 SQL 文件?
- Spark Optimization and Hands on Workshop
-
[Workflow]
-
[Install and Delopyment]
-
[Performance and HA]
-
[Security]
Data On EKS
- EMR on EKS Best Practice Guide
- EMR on EKS workshop
- Mobileye: Spark on EKS migration
- Tool to convert spark-submit to StartJobRun EMR on EKS API
- Orchestrate an Amazon EMR on Amazon EKS Spark job with AWS Step Functions
- data-on-eks
- EMR on EKS 与 Apache Kyuubi 的数据驱动之旅
- Spark on EKS workshop
Stream - Flink and Spark Streaming
- [Flink on EMR]
- 基于 Hudi + Flink多流拼接(大宽表)最佳实践
- ClickStream workshop
- Clickstream Analytics on AWS
- 多库多表场景下使用Amazon EMR CDC实时入湖
Stream - Kinesis
- How to do analysis and virtulization DynamoDB
- AWS Kinesis Workshop
- Sending Data to an Amazon Kinesis Data Firehose Delivery Stream
- lambda as a consumer for kinesis
Stream - Kafka
-
[Connection]
-
[Reliability]
-
[Performance & Cost]
Ad-hoc and Interactive query: Athena
-
How to use the Athena to create the complex embeded table and query the table
-
[Athena Perfomrance]
-
Use CTAS statements with Amazon Athena to reduce cost and improve performance
Data Warehouse: Redshfit
-
[Usage]
-
[Redshift performance]
-
[CDC to Redshift]
-
[ClickHouse and S3]
-
[Streaming datawarehouse]
Search and analytics: Elasticsearch Service
Governance
-
[Lake Formation]
-
[Data Quality with Deequ]
-
[Data Quality with Great Expectations]
-
[Metadata management]
BI
-
- Configure ADFS Identity Federation with Amazon QuickSight
- Enabling Amazon QuickSight federation with Azure AD
- Manage users and group memberships on Amazon QuickSight using SCIM events generated in IAM Identity Center with Azure AD
- AWS Managed Microsoft AD to authenticate users in QuickSight
- QuickSight deployment models for cross-account and cross-Region access to Redshift and RDS
-
Integrate Power BI with Amazon Redshift for insights and analytics
Delta Lake
-[DataBricks]
-
[Hudi]
-
[Iceberg]
IOT
IoT Core
IoT Timeseries
OEE
IoT anaytics
Edge
OTA
AIOT
Security
- AWS Security Hands on Lab - URL need whitelist
- AWS Security Hands on Lab2
- Public Access Consideration
- Curated list of links, references, books videos, tutorials, Exploit, CTFs, Hacking Practices etc. which are related to AWS Security
- An AWS Pentesting tool that lets you use one-liner commands to backdoor an AWS account’s resources
- Top 2022 AWS data protection service and cryptography tool
- BMW automation compliance-at-scale
- Automated Security Response on AWS
- [Security For AI and AI For Security]
Encryption - KMS
- Share-CMK-across-multiple-AWS-accounts
- Using-SM-Key-Algorithm-in-China
- Demystifying KMS keys operations, bring your own key (BYOK), custom key store, and ciphertext portability
- bring your own key to AWS KMS
- Multi-Region keys in AWS KMS
Credential - Secret Manager
Certificate - Certificate Manager
-
How to build a CA hierarchy across multiple AWS accounts and Regions for global organization
-
[Validate the ACM certificate]
Asset Management and Compliance
AuthN and AuthZ
-
Summary the Single-Sign-On cases
- Enabling Federation to AWS console using Windows Active Directory, ADFS, and SAML 2.0
- Using IAM federation and Switch role to implement the Single Sign On multiple AWS Accounts
- Okta-OpenID-AWS-in-the-Browser
- Enabling custom identity broker access to the AWS console
- Grant my Active Directory users access to the API or AWS CLI with AD FS
- Using-temporary-credentials-with-AWS-resources
- Okta - AWS China multi-account console integration
- Keycloak on aws
- Keycloak with Okta OpenID Connect Provider
- Managing temporary elevated access just-in-time access to your AWS environment
- Using global region SSO service to federate China region console
-
[Access Role outside AWS]
Sentitive Data
Threat detection - GuardDuty
WAF
- aws-deployment-with-fortiweb-waf Source Code
- WAF Agent Bot
- AWS WAF-Workshop
- WAF-Simulation-With-DVWA
- 使用 Amazon WAF 进行 Captcha人机验证
- WAF的托管规则说明
- 中国区抗DDoS方案
- Strengthen Your Web Application Defenses with AWS WAF
- AWS WAF AntiDDoS - Global Region
- AWS WAF Rules Reviewer: 一键审查你的 WAF 配置
Permission - IAM Policy, S3 Policy, RAM Policy
Multi accounts structure
SIEM and SOC
- Security Hub quick start
- Customer security findings for security hub
- SIEM on Amazon OpenSearch Service
Vulnerability Assessment - Inspector and Alternative
Network
Test Performance/Latency
VPC
- How to solve private ip exhaustion with private nat solution
- How do I troubleshoot network performance issues between EC2 and on-premises host over the internet gateway
- How do I modify the IPv4 CIDR block of my Amazon VPC
Keep private - VPC Endpoint and PrivateLink
-
Why can’t I connect to an S3 bucket using a gateway VPC endpoint?
-
How do I configure cross-Region Amazon VPC interface endpoints to access AWS PrivateLink resources?
NAT and proxy
Load balancers
Cross data center and cloud Leasing Line - Direct Connect and VPN
-
Amazon Direct Connect inter-region routing for public access resources
-
build IPsec tunnels with static or dynamic customer gateway IP addresses - certificate-based VPN
-
Direct Connect and AWS Local Zones interoperability patterns
Cross board transfer
- Cross region EC2 to EC2 transfering speed testing
- cross-border-data-synchronization with data transfer hub
Cross accounts and Cross VPCs - TGW
Acceleration network
-
Using Amazon Global Accelerator to improve cross board request improvement
-
Measuring AWS Global Accelerator performance and analyzing results
-
Optimizing performance for users in China with Amazon Route 53 and Amazon CloudFront
-
使用 Amazon CloudFront + Amazon S3 + AWS Lambda@Edge 动态调用业务接口生成图片
-
Using CloudFront Origin Shield to protect your origin in a multi-CDN deployment
Edge
Network Secuirty
-
[Traffic Mirror]
DNS
Route 53
HTTPDNS
DNS hijacked using http dns bypass
Serverless
Serverless Workshop
Function as Service - Lambda
-
Lambda integration
- Using AWS Lambda with Amazon Kinesis
- How to put the S3 event to Kafka using lambda
- Demo how to send the Lambda logs to S3 and ElasticSearch by using Kiensis Firehose
- Run the serverless wordpress with AWS Lambda and AWS EFS
- AWS 告警通知到微信
- Lambda write PostgreSQL
- Lambda sent email
- 使用 Lambda 函数URL + CloudFront 实现S3镜像回源
-
Lambda usage
- Schedule-Invoke-Lambda
- AWS Lambda Custom Runtime for PHP
- How to clean up the elastic network interface created by Lambda in VPC mode
- How to get the lambda public IP address
- How to retrieve the System Manager Parameter Store from lambda
- Understanding the Different Ways to Invoke Lambda Functions
- Run web applications on AWS Lambda without changing code
- Disney use the open source and serverless
- Liftshift web app to serverless - part1
- Liftshift web app to serverless - part2
- lambda extensions
- 利用无服务器技术快速构建全球分布式应用 - 将 Lambda URL 作为 CloudFront 的回源地址, 使用 Route 53 和 CloudFront 就近回源请求
-
Lambda cost
-
Lambda performance
API Gateway
Step function
Build the serverless - SAM, Chalice, Serverless framwork, CDK
Serverless with AI/ML
Migration
Journey to Adopt Cloud-Native Architecture
- #1 – Preparing your Applications for Hypergrowth
- #2 – Maximizing System Throughput
- #3 – Improved Resilience and Standardized Observability
- #4 – Governing Security at Scale and IAM Baselining
- #5 – Enhancing Threat Detection, Data Protection, and Incident Response
Active Directory
Database
-
Migrating SQL Server to Amazon RDS using native backup and restore
-
Best practices for migrating PostgreSQL databases to Amazon RDS and Amazon Aurora
-
Aurora launches instances in at least 3 AZ even if less are specified
Data migration tool - DMS
Data migration tool - 3rd party tool
- Migration-Data-From-AliCloud
- XData
- Flink CDC Database Data
- 使用 RisingWave 实现 MSK,Kinesis,RDS MySQL 实时数据同步
Cross Cloud Migration
- Migrate from AliCoud workshop
- Assess secure Windows Servers for TCO analysis using Migration Evaluator
- Resource Discovery for Azure
- AWS AI-Powered Migration Assistant - Cloud Migration assessment
File migration
VMware migration
Accelerating Migration Evaluator discovery for VMware environment
Modernization
Database Modernization
AWS Transform
- Accelerating .NET and SQL Server Modernization using Agentic AI
- [Workshop execution record]
Storage
S3 cross region or cross cloud OSS
-
How to sync S3 bucket data between global region and China region
-
Synchronize S3 bucket contents with Amazon S3 Batch Replication
S3
-
Uploading to Amazon S3 directly from a web or mobile application
-
S3 disale TLS1.1 access or enforce TLS1.2 for in-transit encryption
-
Hosting Internal HTTPS Static Websites with ALB, S3, and PrivateLink
-
How Trend Micro uses Amazon S3 Object Lambda to help keep sensitive data secure
-
通过 STS Session Tags 来对 AWS 资源进行更灵活的权限控制 - 但是需要一个认证机制去确保userid可信的
-
Hosting Internal HTTPS Static Websites with ALB, S3, and PrivateLink
EBS
Storage Gatewway
- storage-gateway-demo and performance testing
- How can I troubleshoot an S3AccessDenied error from my file gateway
- How can I set up a private network connection between a file gateway and Amazon S3
- Resolve an internal error when activating my Storage Gateway
- 手工激活在IDC内网的Storage Gateway
EFS and FSx or other shared file system
-
Amazon FSx for Lustre or Amazon FSx for Windows File Server Workshop
-
Amazon FSx for Windows File Server file share on an Amazon EC2 Linux instance You can mount an Amazon FSx for Windows File Server file share on an Amazon EC2 Linux instance that is either joined to your Active Directory or not joined.
-
Deploying IPFS Cluster using AWS Fargate and Amazon EFS One Zone
Database
RDS
RDS usage
-
Amazon Aurora MySQL Database Quick Start Reference Deployment
-
Use Proxysql for RDS for MySQL or Aurora databases connection pool and Read/Write Split
-
[PostgreSQL Logging]
-
rds-postgresql ERROR: must be loaded via shared_preload_libraries
-
[MySQL 手工分库分表]
RDS Cross region, cross account, data replication and backup
-
Use Amazon RDS Proxy to provide access to RDS databases across AWS accounts
-
DB Snapshot cross region copy and backup cross region replication
RDS upgrade
-
Achieving minimum downtime for major version upgrades in Amazon RDS PostgreSQL
-
How to Migrate from Amazon RDS Aurora or MySQL to Amazon Aurora Serverless
-
Moving to Graviton2 for Amazon RDS and Amazon Aurora databases
RDS Security
RDS Performance
Graph Database
-
How to use the Neptune to Build Your First Graph Application
-
Diagram-as-code using generative AI to build a data model for Amazon Neptune
ElastiCache
Key-Value and Document
DynamoDB
-
[Migration and Replication]
- 如何将我的 DynamoDB 表从一个 AWS 账户迁移到另一个账户
- Streaming Amazon DynamoDB data into a centralized data lake
- 中国区与 Global 区域 DynamoDB 表双向同步
- aws-dynamodb-cross-region-replication
- DynamoDB table initial migration from global to China
- 使用 Lambda 订阅Amazon DynamoDB 变更数据,并传输到Amazon OpenSearch,实现全文检索
- Understanding Amazon DynamoDB latency
- 大规模 DynamoDB 表数据跨账号迁移指南
-
[Security]
-
[Performance]
MongoDB and DocumentDB
Time series
Container
EKS
Serverless or Kubernetes on AWS
EKS networking
-
Exposing Kubernetes Applications via service and ingress resource
-
How do I expose the Kubernetes services running on my Amazon EKS cluster
-
[How do I use multiple CIDR ranges with Amazon EKS]
EKS practice
-
[EKS Managed Group]
- Overview
- Quotas
- Official doc
- Cluster autoscaler Managed node groups are managed using Amazon EC2 Auto Scaling groups, and are compatible with the Cluster Autoscaler. You can deploy the Cluster Autoscaler to your Amazon EKS cluster and configure it to modify your Amazon EC2 Auto Scaling groups.
- Vertical Pod Autoscaler
- Horizontal Pod Autoscaler
DevOps on EKS
-
Install SSM Agent on Amazon EKS worker nodes by using Kubernetes DaemonSet
-
How can I check, scale, delete, or drain my worker nodes on EKS
-
Application first delivery on Kubernetes with Open Application Model
-
EKS Performance
-
[Karpenter]
-
Reduce container startup time on Amazon EKS with Bottlerocket data volume
ECS
-
ECR Sync up from global from China and ECS Service Discovery
-
How can I create an Application Load Balancer and then register Amazon ECS tasks automatically
-
[Amazon ECS firelens]
Fargate
Istio, Envoy, App Mesh, Service discovery
-
Running microservices in Amazon EKS with AWS App Mesh and Kong
-
[EKS and CloudMap]
ECR
DevOps
Management
[AWS Well-Architected]
- AWS Well-Architected labs
- AWS Well-Architected Labs in Chinese
- AWS Well-Architected Agent SKILL
- Resilience and Seucirty - SSR KPI 实操指南(Quick Apps版本 + Desktop版本)
[Organizing Your AWS Environment Using Multiple Accounts]
CI/CD
Serverless CICD
Container CICD
Monitoring and Tracing
-
Monitor using Prometheus and Grafana Here is how to deploy Grafana on EKS
-
Set up cross-region metrics collection for Amazon Managed Service for Prometheus workspaces
-
Create cross-account, custom Amazon Managed Grafana dashboards for Amazon Redshift
-
[Grafana and CloudWatch integration]
Logging
-
[Log Hub]
Change configuration
-
Collecting AWS networking information in large multi-account environments
-
Find Public IPs of Resources – Use AWS Config for Vulnerability Assessment
-
service-screener - evaluate their AWS service configurations
Developer
Infra as Code
Integration
Quque, notification
Call Center
-
Automotive Call Center Services Solution Using Amazon Connect
-
Defending Against SMS Pumping: New AWS Features to Help Combat Artificially Inflated Traffic
-
Reduce risks of user sign-up fraud and SMS pumping with Amazon Cognito user pools
MQ
- AmazonMQ-Workshop
- Automate RabbitMQ configuration in Amazon MQ
- consumer_timeout Amazon RabbitMQ 不支持修改
- [Jenkins 与 RabbitMQ集成]
Media
Video on Demand
Video Streaming
Mobile
Moible app development
GraphQL - AppSync
Business continuity
Backup
DR
-
Understand resiliency patterns and trade-offs to architect efficiently in the cloud
-
[Building a disaster recovery site on AWS for workloads on Google Cloud]
RDS HA/DR
Resilience
- Verify the resilience of your workloads using Chaos Engineering
- 用Q CLI进行智能韧性测试
- [针对多AZ部署的系统,推荐采用 ARC zonal shift /zonal autoshift 进行快速切换,这是经过AWS自身服务进行了充分验证的措施: Application Load Balancer, Network Load Balancer, Amazon EKS, EC2 Auto Scaling Groups]
- [Zonal Shift 不直接作用于单个 EC2 实例,而是通过 EC2 Auto Scaling Group (ASG) 来实现对 EC2 工作负载的 AZ 流量转移]
- [Zonal Autoshift 在上述资源支持的基础上,还必须配置 Practice Run(演练)]
- Using zonal shift with Amazon EC2 Auto Scaling
Game
GameLift
- [unreal engine game server]
SAP
HA/DR
Office and business application
Workspaces - VDI
AI in the office
- minimax office skills
- Frontend Slides creating stunning, animation-rich HTML presentations — from scratch or by converting PowerPoint files
- Beatiful HTML presentation template
Metaverse
Automotive
-
[Software Define Vechile - SDV]
- soafee aws iotfleetwise demo
- 在AWS上构建基于SOAFEE的云原生软件定义汽车实践
- A Cloud-Native Environment for Distributed Automotive Software Development
- Simulating Automotive E/E Architectures in AWS – Part 1: Accelerating the V-Model
- Simulating Automotive E/E Architectures in AWS Part 2: Solution in Action
- Automotive Demo Lab
- 揭秘安卓 AOSP 系统构建提速 50%
- 车载以太网为什么要用SOME/IP?
- How BMW uses AWS to scale and automate SDV with virtual ECUs
- The AWS Architecture Behind BMW Operating System 9: How the Cloud Supports the Latest in Customized In-Car Connectivity
- AWS and Qualcomm Software-Defined Vehicle demonstrator for Cloud-Native Snapdragon Digital Chassis development
- 借助 AWS Graviton 实现 Android Automotive 虚拟化,打造高效虚拟化车载系统
-
[Autonomous Driving]
- ADDF is a collection of modules, deployed using the SeedFarmer orchestration tool. ADDF modules enable users to quickly bootstrap environments
- Develop and deploy a customized workflow using Autonomous Driving Data Framework (ADDF) on AWS
- 详解智能驾驶的功能与场景体系
- Deploy and Visualize ROS Bag Data on AWS
- 在中国区部署 SIMPHERA Architecture
- 理想汽车过去一年发表的16篇智驾及AI应用相关论文
- Accelerating AI/ML development at BMW Group with Amazon SageMaker Studio
- AD Insight Hub
-
[Vechile Data Plantform - VDP]
-
[Connected Mobility Services - CMS]
-
[Smart Cockpit]
-
[Battery and Charging]
HealthCare and Life Science
-
AWS re:Invent 2025: A transformative moment for healthcare and life sciences
-
[Drug Discovery and R&D]
- [CDD and AIDD - Protein Folding and Design]
- AWS Batch Architecture for Protein Folding and Design
- 基于 Alphafold2 一键构建云上高可用蛋白质结构预测平台
- ESM3 - flagship multimodal protein generative model; ESM C - best protein representation learning model
- AI快速生成全新抗体设计图
- 利用 RFdiffusion + ProteinMPNN + ESMFold 自动设计出一批新的 Herceptin 重链 CDR 变体,并筛选出结构最可信、最值得做湿实验验证的候选序列
- Accelerating the Design of Candidate Drugs Using Amazon SageMaker with Nimbus Therapeutics
- [Drug Discovery Workflows]
- [Statistical Compute Environment using R]
- [CDD and AIDD - Protein Folding and Design]
-
[Data Analysis]
- [R&D data lake]
- [Commerical Data lake]
-
[GenAI in Life Sciences]
- Empowering biomedical discovery with AI agents
- Healthcare and Life Sciences Agent Catalog
- Life Sciences Innovation with Agentic AI
- Claude for Life Sciences
- Nexus-AI
- 临床招募 AI Agent Demo 方案概览
- Demo 1: 智能患者-试验匹配 Agent
- Demo 2: 患者预筛选对话 Agent
- Demo 3: 试验方案智能解析 Agent
- Demo 4: 多Agent协作招募工作流
- Demo 5: 语音招募助手 (BidiAgent)
- [Medical Insights Hub]
- [Medical Deep Insights]
-
[LS and HC Compliance]
Retail and ECommerce
快时尚电商行业智能体设计思路与应用实践
- 1. 借助超长上下文窗口分钟级构建智能客服系统
- 2. 借助 LangChain/LangGraph 和 MCP 重塑行业的智能化生态系统
- 3. 借助 Transcribe/Polly 打造新一代智能语音客服,实现媲美人工客服的对话体验
- 4. 借助Amazon AgentCore 实现智能客服 SOP 的可执行 MCP 工具化实践
- 6. 借助 Amazon Bedrock AgentCore MCP Server,Amazon Bedrock,Strands Agents,Kiro 实现智能体极速研发
- 7. Amazon Bedrock AgentCore Runtime 深度解析和场景分析
- 8. 基于 WebSocket 的语音系统:Nova 2 Sonic, AgentCore, Strands Agents 企业级架构实践
- 基于Strands Agents SDK和Amazon Bedrock AgentCore构建商品详情图广告词审查Agent
E-Commerce 选品和设计
- 依托 Amazon Bedrock AgentCore 紫讯打造浏览器自动化智能体 BrowserAct
- 阿尔特携手 Amazon AgentCore ,打造懂你的AI,智能搜索成本降低34%
- 垂直电商以图搜图再升级:DINO 模型带来精准匹配体验
- 图片侵权检测
Amazon Global Selling
导购和客服
- How Amazon is using generative and agentic AI to transform the shopping experience
- Amazon’s next-gen AI assistant Rufus for shopping is now even smarter, more capable, and more helpful
- How Rufus scales conversational shopping experiences to millions of Amazon customers
- Amazon’s new AI-powered shopping feature ‘Help Me Decide’ makes it easy to quickly pick the right product
- Introducing Amazon Lens Live: Instant scanning, real-time product matches, and insights from Amazon’s AI Shopping Assistant
- Visa Intelligent Commerce on AWS: Enabling agentic commerce with Amazon Bedrock AgentCore
物流和供应链
FSI
Banking & Online Banking
撮合
Payment
- Agents that transact: Introducing Amazon Bedrock AgentCore payments, built with Coinbase and Stripe
- Zero-ETL 在支付业务离线数据上的实践
Hospitality
추천 도구
다른 키워드를 입력하거나 필터를 제거해 보세요.
설치
npx skillfish add liangruibupt/aws-is-how