This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass
What I Built
What if your surroundings could give you your next adventure?
NOMAD turns a photograph of your environment into a personalized outdoor exploration mission. Instead of using AI to keep people scrolling, NOMAD uses it to inspire them to put their phones down and explore the world around them.
Users upload a photograph of their surroundings. NOMAD analyzes visible environmental details and generates a short mission with a memorable title, practical steps, and an explanation of what inspired it.
The goal is to make ordinary places feel worth exploring again. NOMAD is designed for students, curious explorers, and anyone who wants to spend less time on screens and more time engaging with the physical world.
The screen helps start the experience. The real adventure happens outside it.
Demo
Try uploading a photograph and see the exploration mission generated from your surroundings.
Code
The project uses React and Vite for the frontend and Python with FastAPI for the backend. The backend validates image uploads and connects to the AI service to generate missions.
How I Built It
I built NOMAD using React, Vite, Python, FastAPI, and the OpenAI-compatible SDK to communicate with the Gemini API. The application is deployed on Render.
The frontend handles the user experience, image uploads, and mission presentation. The backend processes uploaded images and requests an environment-aware mission from the AI model. The mission instructions include practical safety constraints and encourage users to focus on their surroundings rather than their screens.
I also implemented spoken mission instructions and XP-based progress tracking to make exploration feel more interactive.
Open-source AI status: The current deployed version uses the Gemini API. I have not yet integrated an open-weight model into this version, so this implementation does not yet demonstrate the challenge's required open-source AI core.
Why Does Open Innovation Matter?
NOMAD explores a different relationship between AI and everyday life: instead of making people spend more time in an app, it uses AI to encourage real-world experiences.
Open innovation could make NOMAD more adaptable by allowing developers to experiment with open-weight models, change the mission-generation logic, and eventually support different inference setups. It could also create opportunities for community contributions to environmental understanding, accessibility, and safer outdoor activities.
My goal is to make AI a bridge to the world outside the screen—not another destination that keeps people looking at it.
My Agent Session
Optional — add a DevRelay session link if available.
Prize Categories
- Best Use of Render — the application is deployed on Render.
Top comments (0)