The Sixty-Minute Workshop (the kit, pending the class)
The exercise is film three's. An agent builds one study card in Figma. Figma hands back its render and its data. Then a second agent builds the card in HTML twice: once from the picture alone, once from the picture plus the data. Both get scored on twenty-seven properties. In film three, the picture
The Sixty-Minute Workshop (the kit, pending the class)
The exercise is film three's. An agent builds one study card in Figma. Figma hands back its render and its data. Then a second agent builds the card in HTML twice: once from the picture alone, once from the picture plus the data. Both get scored on twenty-seven properties. In film three, the picture
Series: Figma for Educational AI Topic: FIGMA · WORKSHOP
Chapters
0:00 Introduction 0:24 Introduction 2:29 Your Turn
YOUR TURN
Your turn. Paste this into Claude: Here is a workshop kit, then paste the read-me. I have sixty minutes and this many students on their own Figma Education accounts. Check the plan's minute budget and its MCP call budget against my class size, and tell me the one step most likely to overrun. Then check two things yourself. Count the calls one stude
@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
- 2:29 — Your Turn
Transcript
A lightly formatted transcript of the video, with timestamps so you can jump to any section on YouTube.
Introduction
[0:00] Habari, this is Liam in for Bear. Is the workshop an hour of building better question an hour of checking? The build takes 30 seconds. Three terms, a facilitator kit, the plan, the prompts, the scoring sheet, and the budget a person runs a class from. The scoring sheet, film three's 27 properties of the card, each marked right or wrong, and the call budget on education, 200 Figma MCP calls a day in 10 a minute per account. The exercise is
Introduction
[0:25] Film three's. An agent builds one study card in Figma. Figma hands back its render and its data. Then a second agent builds the card in HTML twice, once from the picture alone, once from the picture plus the data. Both get scored on 27 properties. In film three, the picture scored four, the data scored 26. Tonight an agent wrote the kit a facilitator runs that from. Four pages, the minute-by-minute plan, the student prompts, the scoring sheet, and the places people got stuck. No students were involved and no student is pretended. The kit is the half an agent can write. The plan gives the hour to seven steps. Five minutes to check the account. 20 minutes of Figma calls, build the card, render it, read it back. Five minutes for the two HTML builds, then a quarter of an hour to render and score, 10
Minutes to fix the one worst row, and five to tally. The build is the short part, the scoring is the hour. Now the arithmetic from film two's method, not a measurement. One student makes six calls. On their own account, that's five of 200 a day once who am I is exempt. 30 students on one shared login would make 180 with no room to retry, and 30 people on the same step in one minute is 30 against the cap of 10. So, own accounts or it doesn't fit. One thing was measured tonight. An agent did the build step once from film three's saved data. 29 seconds, five turns, 14 cents at list price under Bear's login. It scored 26 of 27, the same as film three, and missed the same row, the button stretch to the card's full width. The data said hug. The
Agent didn't apply it. Nobody rendered. That's why the plan spends a quarter of an hour looking. The stuck points in the kit are the series own, not students. From film one, verified but not upgraded, so still on a starter team and starters three file limit. From film three and again tonight, the build can't see its own render. And from the first experiment, two teams and the wrong one has a view seat. The plan checks the account in the first five minutes for a reason. So, can a class do film three's exercise in an hour? Pending. The kit is written, the build is timed, the arithmetic says own accounts. What's unmeasured is everything the class decides. How many finish the sheet in the quarter hour where they actually stop and what their calls add up to. That waits for a class and for Bear.
Your Turn
[2:30] Your turn, paste this into Claude. Here's a workshop kit, then paste the read me. I have 60 minutes and this many students on their own Figma education accounts. Check the plan's minute budget and its MCP call budget against my class size and tell me the one step most likely to overrun. Then check two things yourself. Count the calls one student makes, is it six? And time the build step on your own machine, is it under a minute? The 60-minute workshop, the kit pending the class at Nick Bear Brown.
