Edge AI · Robotics · Mobile · 2025 – 2026

3LYTRA: An Autonomous Robot Platform with On-Device LLM and Computer Vision

Built a robot platform and iOS/Android app: WebRTC remote driving, an on-device LLM on a Jetson Orin Nano, and YOLO object detection on a Raspberry Pi 5.

Founder and AI Architect · Schilke App Solutions LLC

Randolph Schilke

By , Senior AI Engineer
Updated

Live video over WebRTC
30 fps
Two-way push-to-talk audio
16 kHz
App platforms: iOS and Android
2

Architecture

How it works

  1. 01
    React Native app
    iOS and Android
  2. 02
    Firebase and TURN
    Signaling and NAT traversal
  3. 03
    WebRTC
    Video, audio, controls
  4. 04
    Raspberry Pi 5
    Motors, camera, YOLO
  5. 05
    Jetson Orin Nano
    llama.cpp with CUDA
How the phone reaches the robot

Context

3LYTRA is a personal robot you drive and talk to from your phone over the internet. I built the robot software, its hardware integration, and the companion app, and shipped the app to the App Store and Google Play in January 2026.

What I built

  • WebRTC remote control with end-to-end video, audio, and joystick control over the internet, using aiortc on the robot and react-native-webrtc in the app, with Firebase Realtime Database signaling and Metered TURN for NAT traversal.
  • A real-time drive interface with dual-joystick tank control, live 30 fps MJPEG streaming, a servo gimbal, and bidirectional push-to-talk audio as 16 kHz PCM over WebSocket.
  • BLE and WiFi onboarding: a custom BLE GATT peripheral built with bluezero for secure WiFi provisioning, plus mDNS discovery and token-based auth for local and remote fallback.
  • An on-device LLM: Dockerized llama.cpp with CUDA on a Jetson Orin Nano, behind a FastAPI proxy with persistent per-user memory and a streaming chat endpoint.
  • Computer vision: YOLO object detection and color-histogram tracking that power an interactive Hide and Seek game mode.
3LYTRA app screenshot showing remote control of the robot
The 3LYTRA companion app.

Results

  • Launched on the App Store and Google Play in January 2026.
  • Runs the language model and the vision models on the robot’s own hardware.

Contact

Hiring for ML platform or healthcare AI work?

I'm open to Senior and Staff AI engineering, ML platform, and data science leadership roles, and to healthcare AI consulting. Email is the fastest way to reach me, and I reply personally.

Target roles
Senior or Staff AI Engineer · ML Platform / MLOps Engineer · Data Science Lead
Location
Remote-friendly; based in the Jacksonville, FL area
Work authorization
U.S. citizen
Focus
ML platforms, MLOps, LLMs, and healthcare AI