Travel Booking Agent
An AI travel planning and hotel booking platform built with agents, microservices, RAG, and Zalo automation.
Travel Booking Agent is an archived AI-powered travel planning platform. It combines a React web app, NestJS microservices, an Express AI gateway, PostgreSQL, Redis, ChromaDB memory, Gemini web search, GitHub Models, deep research, and automated hotel communication through Zalo.
Archived Project Notice: This project was discontinued and became inactive over a year ago. As a result, some technologies, dependencies, and APIs may be outdated, deprecated, or no longer functioning. This repository is maintained strictly for archival, historical, and educational reference purposes, serving as an academic showcase of microservice architectures, AI Agent state machines, and RAG systems.
Project Highlights
Microservice and AI-agent architecture for travel planning and booking coordination.
Video & Walkthrough
Timeline
Behind The Project
Overview
The platform helps users research destinations, plan trips, and coordinate accommodation booking with AI-assisted workflows. It includes conversational travel chat, deep research, semantic memory, companion management, and an autonomous booking flow.
Problem
Travel planning and hotel booking involve scattered research, repeated preference collection, manual hotel communication, and state tracking. The project explored how AI agents and microservices could turn those steps into a guided product workflow.
Approach
The system uses a monorepo with a React frontend, NestJS auth/user/trip services, an Express AI service, shared libraries, PostgreSQL, Redis, ChromaDB, and Docker infrastructure. AI features include chat, real-time web search, deep research, RAG memory, and Zalo-based hotel communication.
Gallery
Interested in this project?
Let's build practical edge and AI systems for your workflow.