GiftGuide AI：以 Gemma 提供個人化禮物建議

Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle Submission 💸

🎁 What I Built

I built GiftGuide AI, an AI-powered gift recommendation tool that helps people find thoughtful gift ideas based on:

Recipient
Occasion
Interests
Budget
Relationship

The user enters these details, and Gemma generates four personalized gift recommendations with the gift name, why it fits, and an estimated price.

💡 Why I Built It

Choosing a good gift can be surprisingly difficult, especially when you know someone's interests but don't know what would actually suit them.

I wanted to build a simple tool that turns a few details about a person into practical gift suggestions.

I also wanted to explore how open-source AI can be turned into something that could potentially become a small side hustle.

🤖 How Gemma Is Used

Gemma is at the core of the application.

The user's information is included in a prompt containing:

Recipient
Occasion
Interests
Budget
Relationship

Gemma then generates exactly four personalized recommendations.

Each recommendation includes:

Gift Name

Why it fits

Estimated price

✨ Features

🎁 Personalized gift recommendations
🤖 Gemma-powered AI suggestions
💝 Interest-based recommendations
💰 Budget-aware suggestions
🎉 Occasion-based recommendations
💻 Simple web interface

💼 Who Would Pay for This?

This could potentially be useful for:

Gift shops
Personalized gift businesses
Event and wedding planners
Small businesses offering gift recommendations
People who frequently need personalized gift ideas

A future version could be offered as an AI recommendation service for businesses that want to help customers choose gifts.

🛠️ How I Built It

Frontend

HTML
CSS
JavaScript

Backend

Python
Flask

Gemma
Google GenAI SDK

The application sends the user's preferences to the Gemma model and displays the generated recommendations in the browser.

📸 Screenshots

🌱 Why Open Innovation Matters

Open AI models make it easier for students, developers, and small creators to experiment with AI and build useful applications without having to train an AI model from scratch.

This project helped me explore how an accessible AI model like Gemma can be used as the core of a practical application.

🔗 Project

🚀 Future Improvements

Some ideas for future versions:

Gift shopping links
More detailed price filtering
Better recommendation formatting
Occasion-specific recommendations
Public deployment
Business-focused features

Built for the Hacktoberfest 2026 Week 2 — Side Hustle Challenge.