TouchGrass-AI：以開源 AI 將螢幕時間轉化為戶外冒險

TouchGrass-AI: Turning Screen Time into Real-Life Adventures with Open-Source AI 🌱

What I Built

I built TouchGrass-AI, a web project designed to encourage people to take breaks from their screens and spend more time exploring the real world.

In a world where we spend hours scrolling through our phones and sitting in front of computers, TouchGrass-AI aims to make offline activities more interesting, accessible, and fun.

The idea is simple: instead of spending all your free time online, discover small activities that help you reconnect with nature, explore your surroundings, and enjoy life beyond the screen.

Live Demo: https://saloni2008apk.github.io/TouchGrass-AI/

GitHub Repository: https://github.com/Saloni2008apk/TouchGrass-AI

Why I Built It

As a first-year B.Tech CSE student specializing in AI and ML, I wanted to build something meaningful while learning how technology can solve everyday problems.

Digital devices help us learn, communicate, and work, but excessive screen time can also make us forget about the world around us.

I wanted to turn this problem into an opportunity by building a project that encourages people to take action offline instead of simply consuming more digital content.

My goal is to make going outside feel less like a task and more like an adventure.

How I Built It

I started by building the website using:

HTML for the structure.
CSS for the visual design and styling.
JavaScript for interactivity.

I developed the project using Visual Studio Code and published the website through GitHub Pages.

My next step is to integrate an open-weight AI model or an open-source AI framework to generate personalized outdoor activity suggestions based on a user's interests, available time, and surroundings.

This would help transform TouchGrass-AI from a simple website into a more personalized AI-powered experience.

Where Open-Source AI Comes In

Open innovation matters because I want TouchGrass-AI to be accessible, affordable, and adaptable.

Using an open-weight model or an open-source AI framework can give me more control over how the project works. It can also make it possible to experiment with different models, reduce dependence on paid AI APIs, and explore privacy-friendly approaches.

For example, a future version could use a locally running model to suggest activities without sending a user's personal preferences to an external AI service.

I also want to explore whether a lightweight model could run on an ordinary laptop, making the experience more accessible to students and people with limited budgets.

My goal is to make open-source AI the engine behind useful real-world action, rather than adding AI just for the sake of it.

Transparency note: The current version is a web prototype built with HTML, CSS, and JavaScript. Open-weight model integration is a planned next step, not a feature I am claiming is already implemented.

Why This Could Be a Side Hustle

TouchGrass-AI could grow into a small product for people who want healthier digital habits and more enjoyable offline experiences.

Potential users include:

Students who spend too much time on their phones.
Remote workers who need meaningful screen breaks.
Friends looking for fun, low-cost outdoor activities.
Families and communities interested in offline challenges.

Possible future features include personalized activity plans, daily outdoor challenges, group challenges, and premium themed activity packs.

An open-source AI approach could help keep operating costs manageable and make the product easier to customize for different communities.

The business idea still needs real-user testing to determine which features people find valuable enough to pay for.

What Makes This Project Different?

TouchGrass-AI focuses on a simple but meaningful idea: using technology to help people spend less time using technology.

Rather than encouraging users to stay on a platform for longer, the project aims to turn a digital interaction into an offline experience.

I want the AI to help people discover what they can do in the real world, then give them a reason to close the tab and go outside.

Challenges I Faced

As a beginner in AI and ML, one of my biggest challenges was turning an idea into a working website while learning the development process.

I worked on the project using basic web technologies and learned about building, organizing, and deploying a website.

The next challenge is integrating a genuinely open-source or open-weight AI component and testing whether it provides useful recommendations without making the product expensive or complicated.

What's Next?

My next goals are to:

Integrate an open-weight AI model or an open-source AI framework.
Generate personalized outdoor activity recommendations.
Explore local inference for better privacy and lower recurring costs.
Test the project with real users and collect their feedback.
Discover whether users would pay for personalized challenges or other useful features.

Try It and Share Your Feedback 🌍

I would love for students, developers, and open-source enthusiasts to try TouchGrass-AI and share their suggestions.

What feature would make you use it regularly? Would you prefer daily challenges, personalized suggestions, or group activities with friends?

I am building this project as a student, learning one step at a time, and I hope to turn it into something genuinely useful.

Built with curiosity, powered by learning, and inspired by the world beyond the screen. 🌱