TerraAI: Build with Open-Source AI, Touch Grass, and Reconnect with Nature

S Ehtishamul Haq — Fri, 09 Oct 2026 11:14:01 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

🌿 What I Built

I built TerraAI — Your AI Outdoor Companion, a web application designed to encourage people to spend less time staring at screens and more time experiencing the world around them.

In a world where we spend hours on our devices, TerraAI takes a different approach: instead of keeping users online, it uses technology to inspire them to step outside.

🌱 What TerraAI offers:

AI-powered outdoor missions: Generate ideas for simple outdoor activities using a locally running AI model.
Mission library: Explore activities such as nature walks, mindful breaks, and neighborhood exploration.
Activity timer: Track time dedicated to outdoor activities.
Progress tracking: Earn points, record completed missions, and track outdoor minutes.
Persistent progress: Save statistics and checklist selections in the browser.
AI fallback: Continue exploring built-in missions when the AI service is unavailable.

TerraAI is built for students, people who spend a lot of time on their computers, and anyone looking for a simple way to take meaningful breaks.

The idea behind the project is simple:

Technology should help us experience more of the real world, not just spend more time looking at a screen.

🚀 Demo

🌐 Live Website: https://ehtishamhaque1212-arch.github.io/TerraAI/

Visit TerraAI, explore the available missions, start an activity timer, and track your progress.

Note: AI-generated missions require the local Ollama service to be configured and running. The built-in mission experience is available without it.

💻 Code

🔗 GitHub Repository: https://github.com/ehtishamhaque1212-arch/TerraAI

The project is built with HTML, CSS, and JavaScript, with optional local AI inference through Ollama.

The repository includes the website structure, responsive styling, application logic, mission generation workflow, timer, and browser-based progress storage.

I wanted to build something lightweight, understandable, and easy to extend without requiring a large frontend framework.

🛠️ How I Built It

I built TerraAI using:

HTML5 for the website structure.
CSS3 for the responsive layout and visual design.
JavaScript for mission selection, timers, progress tracking, and application interactions.
Ollama as the local AI runtime.
Llama 3.2 as the example open-weight model for generating outdoor activity suggestions.
Browser localStorage for saving user progress without requiring a cloud database.

The AI workflow is designed to run locally. When a user requests new missions, the application sends a prompt to the Ollama API, asks the model to return structured mission suggestions, and displays the results in the interface.

If the AI request fails, TerraAI falls back to predefined missions so the core experience remains usable.

This project helped me explore how open-weight models and local inference can be incorporated into a practical web application.

🔓 Why Does Open Innovation Matter?

For TerraAI, open innovation matters because the purpose of the project is to make AI useful without making a remote AI service a requirement for the core experience.

Using an open-weight model through local inference offers several advantages:

Greater control: Developers can choose and configure the model they want to run.
Privacy-conscious design: Prompts can be processed locally instead of being sent to a third-party AI provider, when the local setup is configured correctly.
Reduced dependence on external services: The application can use built-in missions even when AI is unavailable.
Learning and experimentation: Developers can inspect the workflow, experiment with prompts, and modify the application to suit their needs.
Accessibility for builders: A simple web application can serve as a starting point for experimenting with local AI.

A closed API could generate activity suggestions, too. However, local inference gives developers more control over where inference happens and how the AI component is configured.

For me, open innovation isn't only about making technology available. It's about giving people the freedom to understand it, experiment with it, and build something meaningful with it.

🤖 My Agent Session

I used AI-assisted development to help plan the application, develop its interface, and work through its implementation.

Agent session: Not published yet.

I plan to share a session link here if I create and publish one using DevRelay.

🏆 Prize Categories

Primary focus: Open-source AI application using local inference.

Potentially relevant partner categories: To be confirmed against the official challenge rules before submission.

🌍 Final Thoughts

TerraAI is my attempt to use AI for something that encourages people to disconnect from their devices and reconnect with the world around them.

This is an evolving project, and there is plenty of room to improve it—from more personalized missions and better activity tracking to a more robust local AI integration.

I'm excited to keep learning, building with open-source AI, and exploring how technology can encourage healthier digital habits.

Less scrolling. More exploring. More time outside. 🌿

Hacktoberfest #OpenSourceAI #TouchGrass #Ollama #Llama #WebDevelopment #BuildInPublic

DEV Community: S Ehtishamul Haq