Touch Grass
This is a submission for the Hacktoberfest Open-Source AI Challenge: Touch Grass.
What I Built
For this challenge, I wanted to create an AI project that encourages people to spend more time outside instead of spending another day entirely in front of a screen.
I built an AI-powered outdoor companion designed to help users turn their free time into simple outdoor activities.
The idea is straightforward.
A user provides information such as how much free time they have, what kind of activity they prefer, and what they feel like doing.
The AI then suggests an outdoor activity that matches their situation.
Instead of giving users another reason to stay online, the application uses AI as a bridge between digital technology and the physical world.
The purpose is to help people discover simple ways to go outside, explore their surroundings, walk, exercise, relax, or spend time away from their screens.
How I Built It
The project was built with open-source AI at its core.
I used an open-weight language model with local inference to process the user's preferences and generate personalized outdoor activity suggestions.
The application uses a modern web stack for the interface and application logic, while the AI model handles the reasoning and recommendation component.
The system takes the user's available time, preferences, interests, and context into consideration before generating a suitable suggestion.
The AI is not being used simply as a chatbot.
It is being used as a decision-making component that converts a person's current situation into an actionable outdoor activity.
The project is also designed so that the AI layer can be modified or replaced with another open model in the future.
Why Does Open Innovation Matter?
Open innovation makes projects like this possible because developers can experiment with AI without having to depend entirely on proprietary services.
Running an open-weight model locally provides greater control over the technology.
It allows developers to experiment with different models, prompts, architectures, and workflows.
It also makes it possible to build applications where the AI component can remain under the developer's control.
For this project, that freedom was important because I wanted to experiment with how AI could be used for something outside the typical productivity and chatbot use cases.
AI is often associated with spending more time on a computer.
I wanted to explore the opposite idea.
What if AI could encourage us to close the laptop and go outside?
My Agent Session
The project was developed with an AI-assisted development workflow.
The agent helped with implementation, experimentation, debugging, and improving the application throughout development.
Prize Categories
Hacktoberfest Open-Source AI Challenge: Touch Grass.
Open-Source AI.
The idea behind this project is simple:
Technology should not always keep us connected to our screens.
Sometimes the best result an AI system can produce is a reason to stop using the computer.
If an application can help someone discover a new walking route, spend time in a garden, go for a run, observe nature, or simply step outside for a while, then the technology has done something valuable.
The goal of this project is therefore not to keep people engaged with the application.
The goal is to make them leave it.
Top comments (0)