Detailed Information on Youtube: https://youtu.be/FGaIxlvWVhw - Step by Step Instruction.
-
GitHub Repository:
-
Developers maintain the source code in a GitHub repository.
-
They use Amplify CLI to manage and deploy the application.
-
-
Amplify CLI:
-
The CLI tool is used to initialize, configure, and deploy AWS Amplify projects.
-
Code changes pushed to GitHub can trigger CI/CD pipelines (if set up), deploying changes to AWS Amplify.
-
-
Amplify Web Hosting:
-
The application’s frontend is hosted using AWS Amplify Hosting, which provides a scalable web hosting solution.
-
Developers deploy the frontend code to Amplify Hosting directly from the CLI or CI/CD pipelines.
-
-
Amazon Cognito for Authentication:
-
Cognito handles user authentication and authorization, allowing secure access to the application.
-
Resort managers log in via the UI at https://portal.bluewaveholiday.net using credentials managed by Cognito.
-
-
Route 53:
- AWS Route 53 manages DNS and provides a custom domain for the application, ensuring users can access the site via https://portal.bluewaveholiday.net.
-
Amazon S3:
-
Stores PDFs, XLSX files, websites, and other datasources.
-
These files can be uploaded and accessed as needed by different parts of the application.
-
-
AWS Lambda:
-
Used for event-driven processing.
-
Automatically triggers to refresh data for Amazon Q (possibly through data ingestion, transformation, or updating search indices).
-
Ensures that the data presented in the application is up-to-date.
-
-
Amazon Q:
-
Serves as the generative AI engine for processing and analyzing data.
-
It utilizes the data from Amazon S3 and other integrated datasources to provide intelligent insights or responses.
-
-
IAM Center for IDP:
-
Manages identities and permissions for accessing Amazon Q.
-
Ensures secure access to the AI functionalities, only allowing authorized personnel or services to interact with the AI.
-
-
Chat UI from Amazon Q:
-
Travel agents interact with the system via a chat interface at https://chat.bluewaveholiday.net.
-
This chat UI is powered by Amazon Q, providing generative AI-driven responses and assistance to the travel agents.
-
-
Code Deployment: Developers push code to GitHub and deploy using Amplify CLI.
-
Frontend Hosting: Amplify Web Hosting serves the frontend application, secured with Cognito authentication.
-
Data Storage: Data such as PDFs and Excel files are stored in Amazon S3.
-
Data Processing: AWS Lambda functions handle event-driven updates to ensure fresh data.
-
AI Integration: Amazon Q processes and analyzes data, utilizing IAM for secure access.
-
User Interaction: Travel agents use the chat UI powered by Amazon Q for their operations.
This workflow ensures a streamlined process from development to deployment, with secure access, efficient data processing, and intelligent AI-driven interactions.
Welcome to the Bluewave Holidays Travel AI Agent project, focusing on planning luxury vacations in the Maldives. This guide will walk you through the setup and usage of an AI-powered Booking Assistant using Amazon Bedrock Agents.
Our AI Booking Assistant helps you plan your dream vacation in the Maldives. 🌴 Dive into vibrant marine life, sail towards fiery sunsets, and bask in paradise. Ready for an unforgettable getaway? Let's get started!
-
Customer Interaction (Step 1)
- Customers Provide Booking Information: Customers interact with the frontend UI provided by a Streamlit application. They input their travel preferences, budget, and other booking details.
-
Frontend Interaction (Step 2)
- Provide Chat Interface and Bedrock Agent Traces: The Streamlit application captures the customer's input and provides a chat interface to facilitate real-time interaction. The input data, along with any relevant Bedrock Agent traces, is sent to the backend for processing.
-
API Gateway (Step 3)
- Amazon API Gateway: The API Gateway receives the HTTP requests from the Streamlit application. It acts as an entry point to the backend services, routing requests to the appropriate processing functions.
-
Initial Processing (Step 4)
-
Process the Text Input: The incoming text input is processed by an AWS Lambda function. This function extracts and formats the data appropriately for further processing by the Bedrock Agent.
-
Bedrock Agent: The Bedrock Agent receives the formatted request and processes it. It uses the OpenAPI JSON file stored in Amazon S3 to understand the request structure and expected response format.
-
-
Request Handling (Step 5)
- Process the Request from Lambda: Another AWS Lambda function handles the detailed processing of the request. This function interacts with various components to generate a response.
-
Model Selection (Step 6)
- Amazon Choice of Model: The Lambda function selects an appropriate AI model from Amazon Bedrock to handle the request. This could involve natural language processing, data retrieval, or decision-making models.
-
Knowledge Base Interaction (Step 7)
- AWS Bedrock Knowledge Base Backed by S3 Content: The Bedrock Agent utilizes a knowledge base, which is backed by data stored in Amazon S3. This knowledge base includes extensive information about Maldives resorts, activities, and travel recommendations.
-
Response Generation and Storage
- Amazon S3 to Store OpenAPI JSON: Responses generated by the Bedrock Agent are stored in Amazon S3. This ensures that the system can track requests and responses for auditing and further analysis.
-
Request: Customer provides budget, preferences, start date, and end date.
-
Process: API Gateway receives the request, Lambda processes it, Bedrock Agent generates a vacation plan using stored knowledge.
-
Response: Generated itinerary ID is sent back to the customer.
-
Request: Customer requests a detailed itinerary using an itinerary ID.
-
Process: API Gateway routes the request, Lambda function fetches and formats the itinerary details.
-
Response: Detailed itinerary is sent back to the customer.
-
Request: Customer provides itinerary ID and accommodation details (name, address, check-in date, check-out date).
-
Process: API Gateway routes the request, Lambda function processes the booking.
-
Response: Booking ID is sent back to the customer, confirming the accommodation.
The Bluewave Holidays Travel AI Agent leverages AWS services to provide a seamless and intelligent travel booking experience. By orchestrating various components, it ensures that customers receive tailored and timely responses, making their vacation planning effortless and enjoyable
POST: Plan a Maldives luxury vacation based on the provided budget and preferences. Description: Takes the user's budget, preferences (e.g., overwater villa, spa, diving, private beach), start date, and end date to plan a vacation, generating an itinerary ID.
GET: Generate a detailed itinerary for a Maldives vacation using the provided itinerary ID. Description: Retrieves and generates a detailed itinerary of the vacation based on the provided itinerary ID.
POST: Send the generated itinerary via email or SMS. Description: Sends the itinerary, identified by its ID, to the specified email address or phone number via the chosen method (email or SMS).
GET: Get available luxury resorts in the Maldives based on user preferences. Description: Retrieves a list of luxury resorts in the Maldives that match the user's specified preferences.
POST: Book luxury accommodation in the Maldives based on the itinerary ID and accommodation details. Description: Books accommodation using the provided itinerary ID and details such as accommodation name, address, check-in date, and check-out date.
GET: Review the details of the Maldives vacation itinerary based on the itinerary ID. Description: Allows users to review the detailed itinerary of their Maldives vacation using the provided itinerary ID. This schema focuses on providing a comprehensive and luxurious vacation planning experience in the Maldives through Bluewave Holidays' AI travel assistant.