The Certification Rubric
The draft has fourteen rows across four sections: reading the design, design decisions the agent can test, the agent writes to Figma and you review, and planning the day. Fourteen rows from the verdicts of films one through thirty-four. Six rows are still pending.
The Certification Rubric
The draft has fourteen rows across four sections: reading the design, design decisions the agent can test, the agent writes to Figma and you review, and planning the day. Fourteen rows from the verdicts of films one through thirty-four. Six rows are still pending.
Series: Figma for Educational AI Topic: FIGMA · CERTIFICATION
Chapters
0:00 Introduction 0:24 Introduction 2:45 Your Turn
YOUR TURN
Your turn. Paste this into Claude: Here is my Figma certification rubric draft. Each row names a skill, a film, and a verdict. For each strong row, write one prompt a student pastes to demonstrate it, and one measurable check they run themselves. For each middle row, name the one constraint that must be resolved before it becomes a strong. Flag any
@NikBearBrown
Part of the Figma for Educational AI series.
Sources and citations: see the reel's SOURCES.md in the open textbook.
▶ The full series playlist — Brutalist https://www.youtube.com/playlist?list=PLG9H-C6rp5RU
What this video covers
- 0:00 — Introduction
- 0:24 — Introduction
- 2:45 — Your Turn
Transcript
A lightly formatted transcript of the video, with timestamps so you can jump to any section on YouTube.
Introduction
[0:00] Ayubowan, this is Liam in for Bear. What does a Figma for educational AI course need to know? Better question, what does it need to show? Three terms, a rubric, what you show, not what you have been told. Each row names the film whose verdict proved the skill teachable. A verdict, strong, middle, or pending. Only strong and middle rows are on the rubric tonight. And pending Bear, this draft waits for his edit and sign-off. The rubric is his.
Introduction
[0:24] The draft has 14 rows across four sections. Reading the design, design decisions the agent can test, the agent writes to Figma and you review, and planning the day. 14 rows from the verdicts of films 1 through 34. Six rows are still pending. Section A, verify the account plan. Some AI features are higher at only film one middle. Redesign data over screenshot. 26 of 27 properties on the first try, film three strong. Variables to a theme switch with no hard-coded hex, film six middle. Component descriptions, an unguessable rule went zero of three to three of three, film seven strong. Notes on frames and variants. Not component sets, the variants note reaches the agent, the sets does not, film 16 strong. Section B, design four uncertainty states as one variable driven component set. Contrast above 4.5, film 12
Strong. Draw an eval board as a fig jam table. Every cell comes back whole, film 21 strong. Turn the table into a Pytest file, 16 cells, 16 breaks caught. Name one probe that passed for the wrong reason, film 21 strong with a hole. Write a consistent design review skill. Three runs, four of six findings agree. Film 24, middle. Section C, agent draw a screen and merge. Walk the checklist. Name the four items that needed a person, film eight middle. Generate a design library from a token file. 15 of 15 variables match, film 26 strong. Capture a live page, 28 of 28 properties. Variables flattened to hex, film 27 middle. Generate an ER diagram from code. Count the fields wrong after one pass, film 28 middle. Section D, one row. Budget the agent's calls before class. From film
Two, eight experiments use 61 of 200 calls, three to 13 per task. A student on their own account can run the biggest task about 15 times in a day. Film two, strong. Six pending rows, Figma first or Figma last. A spec in FigJam persists pros in a chat doesn't. Film 33, pending bear. Student misreads of the uncertainty states, film 22. The version log actually run, film 17 and 32. The audit and continuous integration, film 31. Week over week diagram drift, films 20 and 30. The workshop and student onboarding, films 25 and 34. Each becomes a rubric row when its film settles. Verdict, pending bear. 14 rows tonight. He reads the films, weighs each row, cuts what is not teachable and signs. The pending rows join when their film settle. Until he signs, this is the agents draft.
Your Turn
[2:46] Your turn, paste this into Claude. Here's my Figma certification rubric draft. Each row names a skill, a film, and a verdict. For each strong row, write one prompt a student pastes to demonstrate it in one measurable check they run themselves. For each middle row, name the one constraint that must be resolved before it becomes a strong. Flag any row where the films claim holds for n equals one only. The certification rubric at Nick Bear Brown.
