#7 in our series of 25 Multimodal Demos for 2025: Building with Gemini 2.0 and Pipecat
Aerial Intelligence, from @maxwelllwang
Voice-controlled drones with sophisticated instruction following, navigation, object detection and tracking, and more. Maxwell calls this Jarvis in the sky.
Built with @Google Gemini 2.0, @nvidia Jetson, and @pipecat_ai.
You can hear Maxwell laugh to himself, in the video, when he says "start looking for an old USPS truck."
How much fun is it to live in a future where a multimodal LLM can interpret such an incredibly specific, yet open-ended and unplanned, visual query?
Maxwell Wang, Starlink innovator at @SpaceX, is revolutionizing drones into IoT platforms—flying Linux computers powered by 5G and real-time AI. 🚁
Imagine drones with unlimited range, seamless tracking, and the possibility of Starlink connectivity.
Subscribe to get notified when the episode releases: https://t.co/xkt5WJZSgM
And one last award from the Real-time Voice and Video Hackathon — the special judges' prize for best use of Open Source software (and models) ...
@maxwelllwang's drone pilot project: voice AI + multiple computer vision models + navigation + complex function calling + a hand-built drone/@nvidia Jetson aircraft.
Congratulations @maxwelllwang. It's really, really fun to see the things you create.