Serve Robotics (Nasdaq: SERV) just showed off Maggie at NVIDIA GTC 2026 — a sidewalk delivery robot that can hold a real-time conversation with pedestrians. Not pre-recorded voice lines. Actual conversational AI, processed at the edge.
What Makes Maggie Different
Serve already has 2,000+ robots rolling across the US, delivering for 3,600+ restaurants through Uber Eats and DoorDash. Those bots navigate. Maggie talks. She uses cameras, sensors, and edge AI to detect obstacles and respond to humans instantly — powered by T-Mobile’s 5G Advanced network for ultra-low latency processing. CEO Ali Kashani put it simply: “We’re building robots that don’t just move through the world, but interact with it.”
The API Layer
Maggie runs on NVIDIA Jetson edge AI with T-Mobile’s edge computing APIs handling localized data processing. For developers and restaurant partners, this means real-time telemetry, delivery status, and interaction data flowing through Serve’s commercial deployment platform. Serve also acquired Diligent Robotics earlier this year for $29M, bringing Moxi — an indoor hospital robot with 1.25M completed deliveries — into the same platform. Two form factors, one API surface.
You Might Also Like
- Googles Gemini 3 1 Flash Live Scores 90 8 on Audio Benchmarks Real Time Voice ai Gets Serious
- Anvil Robotics Openarm the 5000 Dual arm Robot kit Already Shipping to Nvidia
- Nvidia Dreamdojo Teaching Robots to Think by Watching 44000 Hours of us
- Stripe Minions 1300 Pull Requests a Week Zero Human Written Code
- Rivians Founder Raised 500m to Build Factory Robots not Humanoids Mind Robotics Explained
Top AI Product 
Leave a comment