I built NatureQuest AI, a web application that uses a locally running open-weight AI model to encourage people to spend more time outdoors and notice the natural world.
Users can choose an activity such as walking, birdwatching, gardening, cycling, or nature observation, select a duration and difficulty, and generate a personalized outdoor quest.
The application also provides a reflection prompt so users can record their experience after completing a quest.
Demo
The application runs locally on my computer.
Screenshots: Add real screenshots of:
The NatureQuest AI home page.
The quest creation form.
The local Gemma model running through Ollama.
Code
GitHub repository: https://github.com/ShubhamHagawane/NatureQuest-AI
The repository contains the React frontend, Express backend, local AI integration, and setup instructions.
How I Built It
I built the frontend using React, Vite, and CSS. The backend uses Node.js and Express to handle quest-generation requests.
For AI generation, I integrated Gemma 2 2B through Ollama, allowing the model to run locally rather than relying on a paid hosted inference API.
The workflow is:
The user chooses an activity, duration, and difficulty.
The React frontend sends the preferences to the Express API.
The backend sends a prompt to the local Gemma model through Ollama.
The response is validated and returned as a structured quest.
The user can complete the quest and save a reflection in browser storage.
I also included sample quests for situations where the local model is unavailable.
Why Does Open Innovation Matter?
Open innovation makes AI experimentation more accessible to students, independent developers, and communities that may not have access to expensive cloud infrastructure.
Using an open-weight model locally gave me an opportunity to learn how to integrate AI into a real application while keeping inference on my own computer after the model is downloaded.
NatureQuest AI explores a simple idea: AI doesn't always need to keep us staring at a screen. It can help us plan an activity that encourages us to step away from technology and experience the world around us.
My Agent Session
I used ChatGPT as a development assistant while building and debugging the application
Prize Categories
My project is submitted for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass.
The project combines a locally running open-weight AI model with an application designed to encourage outdoor activities and real-world engagement.
Top comments (0)