Community project
Europe Embodied hackathonAn autonomous Go2 inspector.
It runs ROS 2 on a Jetson Orin Nano with WendyOS.
By Anirudh Jupudi, Adyansh Gupta, and Manas Reddy Arumalla. Shared with the team’s permission.
From simulation to the real world
The Full Stack Physical AI Platform
Simulate robots and drones on your laptop, ship the same project to real hardware, and run it in the field. Your coding agent can drive every step.
Install the CLI
curl -fsSL https://install.wendy.dev/cli.sh | bashRuns on
vm:go2-sim
Coding agent
Example session: a coding agent is asked to write a patrol app for a Unitree Go2, test it in simulation, and ship it to the real robot. It edits main.py, starts a simulated Go2 in MuJoCo inside a WendyOS virtual machine, and runs the app there while the simulated dog walks its route. After checking the simulation logs, it deploys the same project to the real robot over USB-C, and the robot reports its waypoints back to the session.
For operations teams
Running a site, not building a robot?
Same platform, packaged for operations. No integration project.
01Simulate
Start in simulation, with no hardware on the desk.
Your agent tests on a simulated robot in a WendyOS VM.
What you can simulate today
Unitree Go2
Previewwendy vm create go2-sim --profile go2Unitree G1
Previewwendy vm create g1-sim --profile g1Any Wendy app
Shippedwendy run --device sim
02Ship
Ship the same project to the real machine.
When the simulation passes, your agent points wendy run at the robot instead of the VM.
Your computer
wendy CLI
macOS (Apple Silicon), Windows, or Linux
MacBook Pro
- USB-C
Plug in one cable, with no network setup.
- Local network
Found by name over mDNS.
- Wendy CloudBeta
Devices dial out, so there are no ports to open.
Your device
Wendy Agent
Real quadcopter
03Operate
Everything a robot needs once it leaves your desk.
See what every app is doing, live, from your terminal or your agent.
Telemetrypatrol · cpu
gpu
soc temp
$ wendy device logs# live logs
$ wendy device top# CPU, GPU, memory
04Agents
Your coding agent runs every stage, from simulation to the fleet.
Add the Wendy plugin to Claude Code or Codex, or connect any MCP client.
Connect your agent44 MCP tools
Then run /reload-plugins and start with /wendy-agentic-coding:wendy-install.
Coding agent setup guide01 · Fast
Test every edit on the device.
A Stagefile rebuilds a Python edit in under 200 ms, so changes reach the device in seconds.
- wendy run --watch
02 · Reliable
Output an agent can parse.
Without a terminal, most commands answer in JSON and fail instead of prompting.
- wendy discover --json
03 · Repeatable
One manifest for every deploy.
One
wendy.jsondeploys the same way from your agent, a teammate, or CI.- wendy json validate
04 · Secure
Guardrails the agent works inside.
Provisioned devices accept the agent only over mutual TLS, and every MCP tool is tagged read-only, mutating, or destructive.
- mTLS
- device pinning
- entitlements
05Units
Runs on the hardware you already have.
Flash WendyOS to a supported board, or install Wendy Agent on the Linux computer or Mac you already run.
UnitModelsRuns
- NVIDIA JetsonOrin Nano · AGX Orin · AGX ThorWendyOS
- Raspberry PiPi 5 · Pi 4 · Pi 3WendyOS
- Qualcomm DragonwingIQ-series boardsWendyOS
- Linux computersAny arm64 or x86_64 · Ubuntu 22.04+ · DGX SparkWendy Agent
- Unitree G1Onboard development computerWendy Agent
- Apple Silicon MacNative Swift apps · macOS 15+Wendy AgentBeta
- ESP32 microcontrollersC5 · C6 · C61 · P4 · S3Wendy LitePreview
06Field reports
Built with Wendy.
Europe Embodied hackathonWendy product
Wendy BoxPortable fire watch, with local detection.
Color and thermal cameras in one kit running WendyOS.
Supports your qualified fire watch and existing site procedures.
Established robotics company “If we found Wendy two years ago, we would have saved $4M in product development costs.”
CEO & Co-Founder
Autonomous farm and lawncare equipment
Fortune 50 oil company “I’m testing WendyOS on a Jetson to enable in-field AI to tune old, dumb control loops. I call it Automated Dynamic Local Loop Control.”
Automation Manager
Refinery production improvement
07Operations
Deploy where the work happens.
Teams run perception, inspection, and control on devices far from a data center.
08Proposal
Start in simulation today, and ship to the real robot when it arrives.
Install the CLI
curl -fsSL https://install.wendy.dev/cli.sh | bashFor teams
Planning a pilot or a fleet?
Talk with our engineers about your hardware, security review, and rollout.
- WendyOS
- Free · Apache 2.0
- Wendy Cloud
- Beta