Y
Pybricks AI Debug Workbench

Debug ownership,
not just the robot.

Turn “it doesn’t work” into evidence a child can explain and act on.

QB03 Demo Day · Hao Ying
01 / 问题
0:00–0:35. 开门见山:我做的是一个给孩子用的机器人调试工作台。它不替孩子修机器人,而是帮孩子从“它不动了”走到“我知道下一步测什么”。
Y logo/ Learning
02
The real problem
It doesn’t work.
CODEROBOT PROFILEMOVEMENT + TELEMETRY

The child sees one failure. Underneath it are five hidden systems—and no usable feedback loop.

01 / 问题
0:35–1:10. 孩子不是没兴趣,也不是不聪明。问题是三类信息没有连起来:代码要求它做什么,robot profile 说明这台机器人怎么接,movement 和 telemetry 记录它实际上做了什么。对孩子来说,最后只剩一句“它不动了”。
Y logo/ Context
03

The child sees

A physical robot
with a history.
What was supposed to happen
Which motor actually moved
What changed since the last run
CONTEXT GAP

A generic chatbot sees

One sentence.
“It doesn’t work.”
No robot state
No grounded evidence
01 / 问题
1:10–1:50. 普通聊天框也救不了这个场景:孩子看着一台有历史、有接线、有动作的机器人,AI 只收到一句话。最不会描述上下文的人,偏偏被要求先把上下文讲完整。
Y logo/ Context alignment
04
The product mechanism

The child and AI debug the same robot.

SHARED DEBUGGING CONTEXT
01 / CODE

What was commanded

The exact program the child ran—not a pasted approximation.

02 / ROBOT PROFILE

How this robot is built

Ports, wiring, motor directions, and attachments captured interactively.

03 / MOVEMENT + TELEMETRY

What actually happened

Visible motion, errors, headings, timing, and an inspectable run trace.

02 / 产品
1:50–2:35. 我具体做了什么:App 自动把三类事实放进同一个调试现场——刚运行的 code;第一次互动测端口留下的 robot profile,包括 ports、wiring 和 motor direction;以及真实 movement 与 telemetry。孩子和 AI 终于在看同一台机器人。
Y logo/ Learning loop
05
Bruner, carried into computing by Alan Kay

Move the robot. See the trace. Work with code.

Enactive / Action

Robot moves

The child runs one small physical test.

→
Iconic / Visual

Motion becomes visible

See the robot move and inspect its telemetry trace.

→
Symbolic / Code

Code becomes meaningful

Connect variables and commands to the observed result.

Movement gives code meaning; code gives movement a model.
02 / 产品
2:35–3:15. 这不是把成人 Copilot 涂成儿童色。Bruner 的三种表征在这里有具体对应:enactive 是机器人真实地动;iconic 是孩子看到动作,并看到 telemetry 的可视轨迹;symbolic 是 code、变量和命令。产品把三者连成一个来回循环,而不是让孩子直接跳进抽象代码。
Y logo/ Product philosophy
06
Capability, not dependency

The useful part is the space we leave.

“三十辐共一毂,
当其无,有车之用。”
有之以为利,无之以为用。
工具给足,轮毂中心仍须留空。
CODE
TELEMETRY
ROBOT PROFILE
AI GUIDANCE
判断
留给孩子
“无”不是缺少,而是产品刻意保护的空间
02 / 产品
3:15–3:55. 儿童产品的核心差异在这里。“有”是代码、数据、上下文和 AI;“无”是我们刻意留下的判断空间。AI 不马上给答案、不自动改完代码,只推动孩子提出一个最小测试。
Y logo/ Real robot evidence
07
Previously tested with a real SPIKE Prime Hub

Code, robot reference, and telemetry in one view.

REAL HUB · REAL RUN · SAVED ROBOT PROFILE
Real Hub run with telemetry and saved robot reference
打开无机器人 Demo ↗
02 / 真机证据
3:55–6:55. 先用右下角链接打开无机器人模式,选 90° 转向过头;请评委先预测,再按预期、观察、假设、最小测试走一遍。回来后用这张真机验收记录说明:同一套流程已经接入真实 Hub、机器人档案和 telemetry。
Y logo/ Real AI coaching
08
Actual Coach Chat after real Hub runs

One observation. One evidence-based next test.

Coach chat explaining a missing motor port
Coach chat guiding repeated measurement
02 / 真机证据
7:05–7:35. 这是实际跑 Hub 后的 Coach Chat。左边根据 ENODEV 和已保存的端口映射,只建议改一处;右边运行成功后仍要求一次测量。它的输出单位不是“答案”,而是“下一次有证据的行动”。
Y logo/ Reusable intelligence
09
More than a prompt

A reusable debugging harness.

TRIGGER

Recognize the debugging moment

Route “it doesn’t work” into evidence collection, not instant advice.

→
CONTEXT

Load grounded robot state

Code, telemetry, profile, boundaries, and the child’s observation.

→
METHOD

Protect student ownership

Expected → observed → hypothesis → one smallest useful test.

Reusable skill · explicit boundaries · inspectable evidence · migration-ready method
查看 Pybricks Debug Coach Skill ↗
02 / 产品
7:35–8:10. 所以我做出的不只是一条 prompt,而是一套可复用的调试 harness:识别时机、加载现场、按固定方法保护学生 ownership。它可以跨模型、跨界面迁移。
Y logo/ Proof
10
Why users will use it

A better next move—for children and coaches.

For children

From random guessing to one small test they can understand and run.

For coaches

Less time taking over ports and wiring; more time coaching engineering thinking.

LIVE NOW: coach.buildwithy.net
03 / 为什么会用
8:10–9:05. 直接回答第三问。孩子愿意用,因为它不给一堵答案墙,只给一个能理解、能执行的小测试;教练愿意用,因为它减少环境排错和成人接管,把时间还给真正的工程思维训练。停在这两个用户价值上,不在这一页展开功能清单或验证路线。
Y logo/ Grow with intelligence
11
Compound capability, not dependency

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.

Y
Learn with AI. Keep the judgment.
9:05–9:45. 慢下来收口:短期结果是机器人今天修好;长期价值是孩子以后能自己拆开陌生问题。产品主页留给评委会后查看。9:45–10:00 停住,给主持人切入问答。