Debug ownership,
not just the robot.
Turn “it doesn’t work” into evidence a child can explain and act on.
The child sees one failure. Underneath it are five hidden systems—and no usable feedback loop.
The child sees
with a history.
A generic chatbot sees
The child and AI debug the same robot.
What was commanded
The exact program the child ran—not a pasted approximation.
How this robot is built
Ports, wiring, motor directions, and attachments captured interactively.
What actually happened
Visible motion, errors, headings, timing, and an inspectable run trace.
Move the robot. See the trace. Work with code.
Robot moves
The child runs one small physical test.
Motion becomes visible
See the robot move and inspect its telemetry trace.
Code becomes meaningful
Connect variables and commands to the observed result.
The useful part is the space we leave.
“三十辐共一毂,
当其无,有车之用。”
工具给足,轮毂中心仍须留空。
留给孩子
Code, robot reference, and telemetry in one view.
One observation. One evidence-based next test.
A reusable debugging harness.
Recognize the debugging moment
Route “it doesn’t work” into evidence collection, not instant advice.
Load grounded robot state
Code, telemetry, profile, boundaries, and the child’s observation.
Protect student ownership
Expected → observed → hypothesis → one smallest useful test.
A better next move—for children and coaches.
From random guessing to one small test they can understand and run.
Less time taking over ports and wiring; more time coaching engineering thinking.
The robot gets fixed today.
The child becomes a debugger for life.
AI creates the conditions for the child to see, judge, and act—then leaves the judgment where it belongs.