TouchGrass AI: Turning Open-Source AI into a Real-World Wellness Side Hustle 🌱
What I Built
I upgraded TouchGrass-AI into an AI-powered outdoor activity planner designed to help students spend less time scrolling and more time enjoying the real world.
Users can enter their available time, budget, and interests to receive personalized outdoor activity suggestions.
How I Built It
- HTML, CSS, and JavaScript for the user interface.
- Ollama to run an open-weight language model locally.
- Local inference to generate personalized activity recommendations.
Why Open Innovation Matters
I wanted to explore an AI approach that gives users more control over their data and lets developers experiment with different models.
Local inference can reduce dependence on paid AI APIs and help keep user prompts on the user's own computer. The project also allows the underlying model to be changed as the application evolves.
The Side Hustle
TouchGrass-AI could become a small wellness product offering personalized activity plans, printable outdoor challenges, or group challenges for students.
What I Learned
This project helped me explore how open-weight AI can power a practical product beyond a basic chatbot.
Project: https://saloni2008apk.github.io/TouchGrass-AI/
Source code: https://github.com/Saloni2008apk/TouchGrass-AI
Note: The features and implementation described above should be updated to reflect what is actually working in the submitted version.
Top comments (0)