This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass
What I Built
I built TrailMate AI, an open-source AI-powered outdoor companion that encourages people to spend less time on screens and more time exploring the real world.
TrailMate helps users decide what outdoor activity they can do based on their available time, interests, and surroundings.
For example, a user can enter:
"I have 1 hour this evening and want to relax outside."
TrailMate can suggest activities such as:
- 🌳 Nature walking
- 🐦 Bird watching
- 🌱 Plant observation
- 📸 Nature photography
- 🏃 Walking or jogging
- 🌅 Sunset activities
The main idea is simple:
Use AI to plan the experience, then put the phone away and go outside.
TrailMate is designed for students, developers, families, runners, hikers, and anyone who wants a simple push to spend more time outdoors.
Demo
🎥 Demo Video: [Add your demo video link here]
🌐 Live Demo: [Add your deployed application link here]
Code
💻 GitHub Repository: [Add your GitHub repository link here]
The project is open-source, so developers can inspect the implementation, experiment with the AI model, and extend the application with their own ideas.
How I Built It
TrailMate AI is built around an open-weight AI model, rather than depending entirely on a closed AI API.
Technology Stack
- 🧠 Open-weight AI model – natural-language understanding and outdoor activity planning
- 🐍 Python – AI and backend logic
- ⚡ FastAPI – backend API
- ⚛️ React – frontend
- 🤗 Hugging Face / open-source AI ecosystem – model integration
- 🦙 [Ollama / llama.cpp / Transformers – use the one actually used] – local model execution
- 🐙 GitHub – open-source development
Basic Architecture
User
↓
Outdoor Goal / Question
↓
TrailMate AI
↓
Open-Weight AI Model
↓
Activity Planning
↓
Outdoor Recommendation
↓
User Goes Outside 🌳
The open model is an important part of the application rather than simply being an optional add-on.
The system can interpret the user's request and generate practical outdoor activity suggestions.
Why Does Open Innovation Matter?
Open innovation was important for TrailMate because I wanted the AI experience to be more transparent, customizable, and privacy-friendly.
🔐 Privacy
An outdoor application may involve personal preferences and potentially location-related information.
With local AI inference, the application can be designed so that sensitive information does not have to be sent to a third-party AI provider for every request.
🧠 Model Freedom
Using an open-weight model gives developers the freedom to experiment with different models instead of being locked into a single closed API.
I can replace the model, test different models, and optimize the system for the specific requirements of the application.
🛠️ Customization
The AI behavior can be adapted specifically for outdoor activities.
Instead of building another general-purpose chatbot, TrailMate focuses the AI on helping users discover activities and spend time outside.
📡 Offline Potential
One of the biggest possibilities of open AI is local inference.
A future version of TrailMate could run directly on a laptop or mobile/edge device and continue working in locations with poor or no internet connectivity.
This is especially useful for hiking, camping, and other outdoor activities.
💰 Cost
Local open-weight models can reduce dependence on paid API calls, making experimentation more accessible to students and independent developers.
For me, open innovation means having the freedom to run, modify, experiment with, and build on the AI technology itself.
My Agent Session
I used DevRelay to document the development process and capture the agent-assisted building session.
🔗 DevRelay Agent Session: [Add your DevRelay session link here]
The session shows how the project was developed and how AI-assisted development was used during the implementation.
Prize Categories
- 🧠 Open-Source AI
- 🤖 AI Agents
- 🌱 Touch Grass / Outdoor Experience
- 🔓 Open-Weight Models
What's Next?
I would like to extend TrailMate AI with:
- 🐦 Offline bird-call identification
- 🌱 Plant identification
- 🗺️ Outdoor route planning
- 🌦️ Weather-aware activity recommendations
- 📍 Local nature exploration
- 📱 A fully offline mobile version
- 🧠 Smaller AI models optimized for edge devices
The goal is not to make people spend more time talking to AI.
It is the opposite:
Use AI for a few seconds → get a plan → put the phone away → go outside. 🌱
Final Thought
AI doesn't always have to keep us behind a screen.
Sometimes the best AI experience is the one that helps us close the laptop, put down the phone, and touch grass. 🌿
Top comments (0)