From lab bench to field trial, fast
Plug a Jetson or Raspberry Pi into your laptop, run one command, and your code is on the device with logs streaming back. Wendy gives research teams a mobile-like developer experience for physical AI — so you iterate on real hardware in minutes, not weeks.
▍
Plug a Jetson or Raspberry Pi into your laptop over USB-C, run one command, and watch your app build, deploy, and stream logs back — like iOS development, but for robots.
If your project has a Dockerfile, it runs on the device. Go from an idea on your laptop to inference on real sensors before the meeting ends.
Python, Rust, Swift, C++, TypeScript — bring the tools your lab already uses. Containerized apps mean no cross-compilation gymnastics or golden SD cards.
Declare the capabilities your experiment needs — GPU, cameras, GPIO, I2C, SPI, audio — and Wendy wires them into your container securely.
Devices run fully offline and are discoverable over USB-C or local mDNS. Run trials in the field, the basement, or the flight cage — no Wi-Fi required.
Stream logs and metrics live during an experiment, and capture sensor data on-device to build the datasets your next model needs.
Containerized apps on a versioned, signed OS image mean every run is pinned: same code, same stack, same device state — publishable, repeatable results.
Push signed over-the-air updates to test units in the field. Fix the bug, ship the new model, and keep the trial running without recalling hardware.
The stack that runs your prototype is the stack that ships: the same OS, runtime, and tooling scale from one devkit on your desk to a certified product line.
Discover, name, and task every devkit in the lab from one console — see what is running where, on which version, without a spreadsheet of IP addresses.
“48 hours of continuous, sleep-deprived coding successfully turned into an edge-deployed, cloud-connected autonomous inspector.”
Anirudh Jupudi
EUROPE EMBODIED Robotics Hackathon — autonomous Unitree Go2 facility inspector
A three-person team deployed a full ROS 2 autonomy stack — RTAB-Map SLAM, Nav2, CHAMP gait, YOLOE gauge reading, and LLM control over MCP — on a Jetson Orin Nano booted with WendyOS, riding a Unitree Go2 quadruped.
Read the storyCommon questions
Direct answers about fast hardware iteration, reproducible experiments, and robotics research workflows with WendyOS.
Research teams need fast iteration on real hardware, not only simulations. WendyOS lets teams build, deploy, and stream logs from supported edge devices so experiments can move from a laptop to a Jetson or Raspberry Pi in minutes.
Yes. WendyOS supports containerized applications and multiple languages, including Python, Rust, Swift, C++, and TypeScript. Teams can bring existing models, services, sensor code, and research tools while using Wendy for device deployment and operations.
WendyOS gives teams a versioned operating system, containerized app deployment, logs, hardware access declarations, and repeatable CLI workflows. That makes it easier to reproduce experiments across a bench of devices or field-test hardware.
Install the CLI, plug a Jetson or Raspberry Pi into your laptop, and run your first app in minutes.