→ Adopted as internal standard; requested as AI consultant by multiple teams post-session
Most designers treat AI as a random visual generator. They type in a guess and hope for something useful. That creates a cycle of trial and error that wastes time and makes it hard to use AI for real product work. I wanted to move the team from exploring to executing by giving them a repeatable way to talk to the AI.
I presented to the design org to make that shift official. Instead of treating AI as a random generator, I showed how it works as a validation tool when you approach it the right way. I introduced the TC-EBC framework: Task (what the screen does), Context (where it fits), Elements (components), Behavior (interactions), Constraints (limits). Every prompt becomes complete and tied to product goals, replacing trial and error with something repeatable.
Configured our internal chatbot as a prompt translator, turning messy design notes into complete TC-EBC frameworks automatically.
Shifted focus from "perfect writing" to core functional questions, producing a copy-paste-ready Figma Make prompt every time.
Used the initial AI output to validate the workflow's logic first, before touching visual iteration.
Established a "3-Iteration Rule" to move from a validated flow to a stakeholder-ready prototype in three focused cycles.
For any Figma Make prompt or design workflow request I share, always do the following:

I presented this live to the whole design org. A structured prompt plus a disciplined iteration process gets a team to alignment much faster than the usual trial and error.
It changed how the team approaches AI: less random experimentation, more repeatable process. The TC-EBC framework became the internal standard for Figma AI work, and Figma Make went from a random visual generator to a validation tool people actually trust.
A few product teams reached out on their own afterward to bring the methodology into their process. Turns out structured prompt engineering is really a design systems problem, and it works a lot better once you treat it like one.