This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine
What I Built
I built SpiritAI, a local-first AI assistant designed to bring the power of artificial intelligence directly to your own machine.
The idea behind SpiritAI is simple:
What if you could have your own AI assistant without depending on someone else's servers?
Most AI assistants rely on cloud APIs, external services, and internet connectivity. I wanted to explore a different approach: an AI assistant that runs using open-weight models locally.
SpiritAI provides a chat interface where users can interact with locally running AI models, manage conversations, and communicate with their AI through a custom-built interface.
It is my interpretation of Ghost in the Machine: an intelligent presence living inside your computer rather than somewhere in the cloud.
Demo
🚧 Project demo coming soon!
I'll share a demonstration video showing:
- The SpiritAI interface.
- Connecting to locally running AI models.
- Downloading models through Ollama.
- Having conversations with a local AI assistant.
Code
GitHub Repository: [INSERT YOUR GITHUB REPOSITORY LINK]
The project is built with open-source technologies and is being developed with local execution as a core principle.
Contributions, suggestions, and feedback are welcome!
How I Built It
SpiritAI is built around local AI inference rather than relying on a proprietary cloud AI API.
Here is the technology behind it:
- Ollama: Runs open-weight AI models locally.
- Go (Gin): Backend API for handling application requests.
- Python (Flask): Supporting backend functionality.
- React: Frontend chat interface.
- JavaScript: API communication and frontend integration.
The application communicates with its backend, which integrates with Ollama to manage model operations and generate AI responses.
Users can pull models and interact with them through the SpiritAI interface.
The important architectural decision was making local inference the foundation of the application, rather than treating it as an optional feature.
Why Does Open Innovation Matter?
Open innovation is the reason SpiritAI can exist in its current form.
Closed AI APIs are convenient, but they introduce dependencies:
- External servers.
- API costs and usage limits.
- Internet connectivity.
- Less control over where conversations are processed.
Open-weight models and local inference give developers another option.
With Ollama, SpiritAI can run AI models on a user's own hardware. Once the required model is downloaded, inference can happen locally without sending ordinary chat requests to a third-party AI provider.
This means users have more control over their data, their models, and their AI experience.
It also means developers can experiment, customize, and build on top of existing open technology without having to create an entire AI model from scratch.
Open innovation makes AI something people can own, experiment with, and build upon — not just something they rent.
My Agent Session
Coming soon!
I'll share my development session and the process behind building SpiritAI.
The Spirit of It
There is something fascinating about opening an application on your computer and realizing that an AI assistant is running entirely on your own machine.
No external AI server needs to answer your ordinary chat requests. No proprietary API needs to sit between you and your assistant.
Just your computer, an open-weight model, and an intelligent system responding to you.
That is the spirit behind SpiritAI.
An AI doesn't always need to live in the cloud. Sometimes, it just needs a home on your machine.
Built with open-source technology. Powered by local AI. Inspired by the Ghost in the Machine.
Top comments (0)