I have been developing an open source multi-modal AI assistant for the past few years that works with any model (local and cloud-based). It does tool calling and receives various inputs such as sensors, camera, voice, etc. I would like to get community feedback on use cases I can further work and integrate.