AI Travel Planner

AI Travel Planner is a web application that generates personalized travel itineraries with the support of Generative AI.

Users select a destination, start date, end date, and travel experience. The application creates a structured day-by-day itinerary that can be viewed, saved, and shared through a link.

The project also demonstrates a complete DevOps workflow with microservices, containerized development, CI/CD, infrastructure automation, deployment, and monitoring.

Main Features

The current application supports:

• destination input
• start and end date selection
• travel experience selection
• AI-generated day-by-day itineraries
• user registration and login
• saving trips
• sharing trips through a generated link
• mock and external LLM providers
• local execution with Docker Compose
• API routing through NGINX
• Prometheus monitoring
• Grafana dashboards
• Kubernetes deployment with Helm
• Azure automation with Terraform and Ansible
• GitHub Actions CI/CD

The GenAI API supports optional budget information, but budget is not yet integrated into the complete frontend and backend workflow.

Architecture

The application consists of:

• React Client
• NGINX API Gateway
• Auth Service
• Trip Service
• Common backend module
• GenAI Service
• PostgreSQL
• Prometheus
• Grafana

Browser
   |
   v
React Client
   |
   v
NGINX API Gateway
   |
   +----------------+----------------+----------------+
   |                |                |
   v                v                v
Auth Service    Trip Service    GenAI Service
   |                |                |
   +--------+-------+                |
            |                        v
            v                 External LLM Provider
       PostgreSQL

For the detailed architecture, see System Overview.

Repository Structure

.
├── .github/
│   └── workflows/          # GitHub Actions workflows
├── api/                    # API contracts
├── backend/
│   ├── auth-service/       # Authentication service
│   ├── trip-service/       # Trip service
│   └── common/             # Shared backend module
├── client/                 # React frontend
├── docs/                   # Project documentation
├── genai/                  # FastAPI GenAI service
└── infra/
    ├── ansible/            # VM configuration
    ├── helm/               # Kubernetes deployment
    ├── monitoring/         # Prometheus and Grafana
    ├── nginx/              # API Gateway
    ├── terraform/          # Azure infrastructure
    └── docker-compose.yml  # Local stack

Run Locally

Prerequisites

Install:

• Git
• Docker
• Docker Compose plugin

Clone the Repository

git clone https://github.com/AET-DevOps26/team-coffeelovers.git
cd team-coffeelovers

Configure the Environment

Create:

infra/.env

Configure the required local values.

For local GenAI development, the mock provider can be used:

GENAI_PROVIDER=mock

Do not commit environment files, API keys, passwords, tokens, or other secrets.

Start the Application

docker compose \
  --env-file infra/.env \
  -f infra/docker-compose.yml \
  up --build

For the complete first-time setup, see How to Launch.

For Docker Compose services, monitoring, logs, and troubleshooting, see Infrastructure Setup.

Local URLs

Component	URL
Frontend	http://localhost:3000
Grafana	http://localhost:3001
GenAI Swagger UI	http://localhost:8001/docs
API Gateway	http://localhost:8080
Auth Service	http://localhost:8081
Trip Service	http://localhost:8082
Prometheus	http://localhost:9090

Use the API Gateway for normal application requests.

Main User Workflow

1. Open the frontend.
2. Register or log in.
3. Enter a destination.
4. Select start and end dates.
5. Select a travel experience.
6. Generate an itinerary.
7. Save the trip.
8. Share the trip through its generated link.

The frontend currently does not collect budget information.

Deployment and Monitoring

The project supports:

• Docker Compose for local development
• GitHub Actions for CI/CD
• Helm for Kubernetes deployment
• Terraform and Ansible for Azure automation
• Prometheus for metrics collection
• Grafana for metrics visualization

Detailed instructions are maintained in the related documentation.

Documentation

Document	Purpose
Problem Statement	Product problem, scope, and users
System Overview	Architecture and service communication
Product Backlog	Implemented and planned functionality
How to Launch	Local launch guide
Infrastructure Setup	Docker Compose, monitoring, logs, and troubleshooting
GenAI Service	GenAI providers, endpoints, and configuration
Kubernetes Deployment	Helm deployment
Infrastructure Automation	Terraform and Ansible
Contributing Guide	Development workflow and contribution rules

Development Workflow

main
  ↑
develop
  ↑
feature/*, bugfix/*, docs/*, hotfix/*

General rules:

• create a branch for every issue
• branch from the latest develop
• open pull requests into develop
• run relevant tests before merging
• update documentation when behavior changes
• do not commit secrets

See Contributing Guide for the complete workflow.

Team Responsibilities

• Paulina: Frontend development and client-side tests
• Adnan: Spring Boot backend services and APIs
• Berfin: GenAI integration, providers, prompts, and quality checks