Guides
Guides · DeviceNexus
The Physical AI Learning Path
Five stages from knowing nothing about robotics to running AI-trained policies on real hardware. Follow in order — each stage builds on the last.
01
What Is a Robot?
What are the physical components of a robot arm, and how do we describe its position in space?
02
How Does a Robot Perceive?
How does a robot sense its environment — cameras, depth sensors, and building maps?
03
How Does a Robot Move?
Given a goal position, how does the robot plan a path and execute it?
04
How Does a Robot Learn? The Paradigm Shift
Instead of programming every motion, how do we teach a robot by showing it? This is where modern robotics diverges from classical automation.
05
How Does a Robot Improve?
Once deployed, how does a robot fleet get better over time — without touching the hardware?
Engineering Deep Dives
3 Numbers That Tell You How to Optimize
Before you try quantization — compute R₁, R₂, R₃ from your model spec. Validated on Jetson Orin and Thor.
Teaching a Robot to Pick Up a Candy Bar
56 demos, 3 training runs, every mistake in between — the full Stage 1→5 pipeline as a single project.
Coming soon