✈️ Safariz — AI-Powered Travel Planner

An intelligent travel planning application that generates personalized day-by-day itineraries using Google Gemini AI, with dynamic images from Unsplash and persistent storage in MongoDB Atlas.

🌐 Live Demo: https://safariz.vercel.app 📦 GitHub: https://github.com/peyalhasan/Safariz

---

🚀 Features

• 🤖 AI-Powered Plans — Generate detailed travel itineraries using Google Gemini API
• 🗺️ Day-by-Day Itinerary — Scheduled activities with times, descriptions, and tips
• 📸 Dynamic Images — Destination photos fetched from Unsplash API
• 💾 Smart Caching — Plans saved to MongoDB Atlas to avoid regenerating the same trip
• 🔗 Shareable URLs — Slug-based URLs like /plan/thailand-tour-4-days
• 📘 Facebook Share — Share plans with Open Graph meta tags
• 📱 Responsive Design — Works on mobile and desktop
• 🔍 Destination Highlights — Key landmarks, food, travel tips, and budget info

---

🛠️ Tech Stack

Technology	Usage
Next.js 14 (App Router)	Frontend & Backend
MongoDB Atlas	Cloud Database
Google Gemini API	AI Plan Generation
Unsplash API	Travel Photos
Tailwind CSS	Styling
Mongoose	MongoDB ODM
Lucide React	Icons
Vercel	Deployment

---

📁 Project Structure

safariz/
├── app/
│   ├── api/
│   │   ├── generate-plan/    # POST: Generate travel plan
│   │   └── plans/            # GET: Fetch latest plans
│   ├── components/
│   │   ├── plan/             # Plan page components
│   │   └── prompts/          # Home page components
│   ├── plan/[slug]/          # Dynamic plan detail page
│   ├── not-found.js          # 404 page
│   ├── error.jsx             # Error page
│   ├── layout.js
│   └── page.js               # Home page
├── lib/
│   ├── gemini.js             # Gemini AI functions
│   ├── mongodb.js            # DB connection
│   └── unsplash.js           # Image fetching
├── models/
│   └── plan.js               # Mongoose schema
├── utils/
│   └── index.js              # Utility functions
└── .env                      # Environment variables

---

⚙️ Getting Started

1. Clone the repository

git clone https://github.com/peyalhasan/Safariz.git
cd safariz

2. Install dependencies

npm install

3. Set up environment variables

Create a .env file in the root directory:

MONGODB_URI=mongodb+srv://<username>:<password>@cluster.mongodb.net/travel-planner
GEMINI_API_KEY=your_gemini_api_key
UNSPLASH_ACCESS_KEY=your_unsplash_access_key
NEXT_PUBLIC_BASE_URL=http://localhost:3000

4. Run the development server

npm run dev

Visit http://localhost:3000

---

🔑 API Keys Setup

MongoDB Atlas

1. Go to mongodb.com/atlas
2. Create a free M0 cluster
3. Add database user and whitelist IP 0.0.0.0/0
4. Copy the connection string

Google Gemini API

1. Go to aistudio.google.com
2. Create a new project
3. Generate an API key

Unsplash API

1. Go to unsplash.com/developers
2. Create a new application
3. Copy the Access Key

---

🔄 How It Works

User enters prompt
       ↓
Check if travel-related (Gemini)
       ↓
Extract destination & days (Gemini)
       ↓
Check MongoDB for existing plan
       ↓
If not found → Generate plan (Gemini)
       ↓
Fetch images from Unsplash
       ↓
Save to MongoDB Atlas
       ↓
Redirect to /plan/{slug}

---

📡 API Endpoints

POST /api/generate-plan

Generate a new travel plan from a prompt.

Request Body:

{
  "prompt": "3 days trip in Thailand"
}

Response:

{
  "slug": "thailand-tour-3-days"
}

GET /api/plans

Fetch the latest 4 generated travel plans.

Response:

{
  "plans": [...]
}

---

🚀 Deployment

This project is deployed on Vercel.

1. Push to GitHub
2. Import project on vercel.com
3. Add environment variables in Vercel dashboard
4. Update NEXT_PUBLIC_BASE_URL to your deployed URL

---

📄 License

This project was built as an assignment for Learn With Sumit.

---

Made with ❤️ by Peyal Hasan