muted cairn
Hey everyone! I'm an indie developer and creator of an open-source project called HomeGenie.
While it started years ago as a Home Automation server, it has now evolved into something more like a portable AI platform with an integrated development environment.
I'm not here to spam links, but to genuinely share a project I'm passionate about and maybe spark a discussion with fellow devs interested in Local AI, IoT, and Computer Vision. The platform is highly extensible and can be programmed directly from the browser using C#, JavaScript, or Python.
The core idea is to provide an "unzip-and-run" AI sandbox. No complex setups, no Python dependencies, no Docker nightmares. It's a lightweight ZIP file that gives you a full dev environment in seconds.
It's built on a hybrid AI philosophy:
🧠 To Reason & Act (Local or Cloud): A unified context engine that translates natural language into real-world system actions. You can power it 100% locally with GGUF models (via our Lailama engine) OR through the cloud using Gemini.
👁️ To See & Interpret (Local): A native Computer Vision subsystem (VisionCore) that runs multi-stream YOLO on a standard CPU.
I recently put together short video demos showcasing these AI capabilities in action, which you can find by searching for the project if you're curious.
I'm really curious to hear your thoughts on this approach to a local-first, portable AI orchestrator. Is this something you'd find useful? Thanks for any feedback!