[SEO Subhead]
This Guidance demonstrates how to create a compelling, AI-driven shopping experience using HAQM Bedrock Agents. It shows how to securely integrate your existing systems and data, minimizing the time and complexity required to build sophisticated generative AI assistants. By following this Guidance, you can enhance your customers’ online shopping experience, providing personalized recommendations, intelligent search capabilities, and engaging conversational interactions.
Note: [Disclaimer]
Architecture Diagram
[Architecture diagram description]
Step 1
A user provides natural language queries to find products or place orders through the application interface.
Step 2
The application is hosted using serverless containers on AWS Fargate, HAQM Elastic Container Registry (HAQM ECR) for managing container images, Elastic Load Balancing (ELB) for traffic distribution, HAQM Route 53 for DNS, and HAQM Cognito for authentication.
Step 3
Static assets, such as product images, are stored in HAQM Simple Storage Service (HAQM S3) with HAQM CloudFront distribution to provide low latency content delivery for an enhanced user experience.
Step 4
The user query is sent to HAQM Bedrock Agents using an HAQM Bedrock API.
Step 5
HAQM Bedrock Agents interprets user input using conversation history, agent instructions, and other configurations, in addition to the underlying foundation model (FM) offered by HAQM Bedrock for contextual understanding of the request, resulting in accurate and relevant responses.
Step 6
The FM uses the given agent instructions including the role of the assistant, specific tasks, step-by-step instructions for completing tasks, and examples, to orchestrate actions for assisting the user in finding and purchasing products.
Step 7
During orchestration, HAQM Bedrock Agents uses ReAct prompting with the Orchestration prompt to invoke relevant actions for finding products or placing orders.
Step 8
To find relevant products from the catalog, HAQM Bedrock Knowledge Bases offers fully managed retrieval augmented generation (RAG) for HAQM Bedrock Agents. It uses product documents stored in HAQM S3 as a data source for embedding product information and HAQM OpenSearch Serverless for vector store and semantic search.
Step 9
To manage API operations such as order creation, inventory checks, and sending email confirmations, HAQM Bedrock Agents uses an Order action group built using an AWS Lambda function and the OpenAPI schema.
Step 10
Using Advanced prompts for Orchestration, HAQM Bedrock Agents is configured with templates for formatting email confirmations and output responses to users for clear and consistent communication.
Step 11
The HAQM Bedrock Agents reasoning process continues until the agent finds relevant products, creates an order with email confirmation, or wants to request additional details from the customer. The final answer is sent back to the application, providing a seamless shopping experience for the user.
Step 1
A user provides natural language queries to find products or place orders through the application interface.
Step 2
The application is hosted using serverless containers on AWS Fargate, HAQM Elastic Container Registry (HAQM ECR) for managing container images, Elastic Load Balancing (ELB) for traffic distribution, HAQM Route 53 for DNS, and HAQM Cognito for authentication.
Step 3
Static assets, such as product images, are stored in HAQM Simple Storage Service (HAQM S3) with HAQM CloudFront distribution to provide low latency content delivery for an enhanced user experience.
Step 4
The user query is sent to HAQM Bedrock Agents using an HAQM Bedrock API.
Step 5
HAQM Bedrock Agents interprets user input using conversation history, agent instructions, and other configurations, in addition to the underlying foundation model (FM) offered by HAQM Bedrock for contextual understanding of the request, resulting in accurate and relevant responses.
Step 6
The FM uses the given agent instructions including the role of the assistant, specific tasks, step-by-step instructions for completing tasks, and examples, to orchestrate actions for assisting the user in finding and purchasing products.
Step 7
During orchestration, HAQM Bedrock Agents uses ReAct prompting with the Orchestration prompt to invoke relevant actions for finding products or placing orders.
Step 8
To find relevant products from the catalog, HAQM Bedrock Knowledge Bases offers fully managed retrieval augmented generation (RAG) for HAQM Bedrock Agents. It uses product documents stored in HAQM S3 as a data source for embedding product information and HAQM OpenSearch Serverless for vector store and semantic search.
Step 9
To manage API operations such as order creation, inventory checks, and sending email confirmations, HAQM Bedrock Agents uses an Order action group built using an AWS Lambda function and the OpenAPI schema.
Step 10
Using Advanced prompts for Orchestration, HAQM Bedrock Agents is configured with templates for formatting email confirmations and output responses to users for clear and consistent communication.
Step 11
The HAQM Bedrock Agents reasoning process continues until the agent finds relevant products, creates an order with email confirmation, or wants to request additional details from the customer. The final answer is sent back to the application, providing a seamless shopping experience for the user.
Get Started
Deploy this Guidance
Well-Architected Pillars
The AWS Well-Architected Framework helps you understand the pros and cons of the decisions you make when building systems in the cloud. The six pillars of the Framework allow you to learn architectural best practices for designing and operating reliable, secure, efficient, cost-effective, and sustainable systems. Using the AWS Well-Architected Tool, available at no charge in the AWS Management Console, you can review your workloads against these best practices by answering a set of questions for each pillar.
The architecture diagram above is an example of a Solution created with Well-Architected best practices in mind. To be fully Well-Architected, you should follow as many Well-Architected best practices as possible.
-
Operational ExcellenceRead the Operational Excellence whitepaper
AWS Cloud Development Kit (AWS CDK) automates and streamlines your entire deployment process for consistent and repeatable rollouts across environments. Gain visibility into user and service actions with AWS CloudTrail, and HAQM Bedrock invocation logging collects comprehensive request, response, and metadata details for insightful troubleshooting and improvement opportunities. These capabilities empower you to reduce deployment time and runtime issues while maintaining consistent environments.
-
SecurityRead the Security whitepaper
HAQM Cognito handles user authentication and authorization for the web app. CloudFront delivers static assets securely with low latency, traffic encryption, and access controls. AWS Identity and Access Management (IAM) policies are scoped down to minimize permissions. The virtual private cloud (VPC) isolates resources in a secure network environment. HAQM Virtual Private Cloud (HAQM VPC) endpoints enable direct, private connections between the VPC and AWS services without internet access, enhancing security. Anonymous user interactions are prohibited, and AWS Shield Standard defends against distributed denial of service (DDoS) attacks.
-
ReliabilityRead the Reliability whitepaper
HAQM Bedrock and Lambda are deployed across multiple Availability Zones (AZs) by default, providing redundancy and high availability. HAQM S3 provides durable storage, and Lambda handles the scaling of your compute resources seamlessly. Route 53 will route your end users to healthy application endpoints through automatic failover, latency-based routing, and health checks, for reliable service delivery.
-
Performance EfficiencyRead the Performance Efficiency whitepaper
Fargate automatically handles scaling and infrastructure management, helping you maximize resource utilization. The Vector Engine for HAQM OpenSearch Serverless provides high-performing vector storage and search capabilities to build generative AI applications. Additionally, CloudFront, a global content delivery network that caches content closer to your users, provides low latency and high transfer speeds.
-
Cost OptimizationRead the Cost Optimization whitepaper
Optimize your costs with Fargate, which charges only for the virtual CPU and memory resources consumed by your containers, eliminating the need to provision and manage infrastructure. HAQM Bedrock provides comprehensive, flexible pricing options to meet your specific requirements. Serve your static content directly from HAQM S3, and use CloudFront to cache and compress content, minimizing data transfer costs and automatically scaling to handle traffic spikes.
-
SustainabilityRead the Sustainability whitepaper
Reduce your environmental impact by using serverless services like Fargate, Lambda, HAQM S3, and HAQM Bedrock, which eliminate the need for server management and maximize resource utilization. These services are built on the energy-efficient infrastructure of AWS, which runs on renewable energy.
Related Content
[Title]
Disclaimer
The sample code; software libraries; command line tools; proofs of concept; templates; or other related technology (including any of the foregoing that are provided by our personnel) is provided to you as AWS Content under the AWS Customer Agreement, or the relevant written agreement between you and AWS (whichever applies). You should not use this AWS Content in your production accounts, or on production or other critical data. You are responsible for testing, securing, and optimizing the AWS Content, such as sample code, as appropriate for production grade use based on your specific quality control practices and standards. Deploying AWS Content may incur AWS charges for creating or using AWS chargeable resources, such as running HAQM EC2 instances or using HAQM S3 storage.
References to third-party services or organizations in this Guidance do not imply an endorsement, sponsorship, or affiliation between HAQM or AWS and the third party. Guidance from AWS is a technical starting point, and you can customize your integration with third-party services when you deploy the architecture.