Skip to main content
Skip to content
Menu

UX Design Intermediate 4 parts Updated

Design a brand-locked AI prompt system

You'll have a written set of AI system prompts and component libraries that force your design tool to output work matching your brand, not generic defaults.

What you'll make

Brand-Locked AI Design Prompt System

  1. Brand Constraints: the 5-7 non-negotiable visual rules
  2. Component Prompts: reusable blocks for layouts, colors, typography
  3. Override Triggers: when to reject AI output and what to fix
  4. Tool-Specific Adaptations: how to modify prompts for Midjourney vs Figma AI vs others

Write each prompt with its use case and paste examples of good vs bad output it produces

The 4 parts

Each part ends with a check you can see.

  1. Audit Your Brand Breaking Points

    Test AI design tools on your current brand and document exactly where they go wrong

    Done when

    You have 8-10 examples of AI design output with specific annotations of what's wrong and what should be different

  2. Build Your Constraint Architecture

    Convert your brand violations into a system of explicit rules that AI tools can follow

    Done when

    You have 5-7 concrete brand constraints written as AI-readable rules with before/after examples

  3. Build Reusable Prompt Components

    Create modular prompt blocks you can mix and match for different design contexts

    Done when

    You have 4-6 component prompts tested for different design needs (headers, layouts, color schemes, typography) that you can combine

  4. Finalize Your Brand-Locked System

    Compile your tested constraints and components into a complete prompt system with tool adaptations

    Done when

    You have a documented prompt system with examples that works across at least 2 different AI design tools

    Capstone: you put together your Brand-Locked AI Design Prompt System.

Where this helps

An elective: a single lesson that sits outside rtwo's job certificates and doesn't count toward one.

About this lesson

What you'll be able to do

  • Build a set of AI prompts that consistently output brand-specific designs, not generic variations
  • Identify where your brand breaks down in AI-generated output and create override rules
  • Have a working prompt library you can copy-paste into any design AI tool
  • Know when to trust AI design output vs when your brand judgment must override it

Who it's for

  • Designers working solo or on small teams who use AI design tools but get frustrated with off-brand output
  • Brand managers who need to control AI-generated visual assets
  • Anyone who's tried Midjourney or similar tools and gotten beautiful but wrong results for their brand

What it won't cover

  • Building custom AI tools from scratch like Stripe did
  • Advanced prompt engineering theory
  • Brand strategy fundamentals — assumes you know what your brand should look like
  • Production deployment or team sharing systems

Skills you'll practise

  • Brand constraint documentation
  • AI output evaluation against brand standards
  • Iterative prompt refinement
  • Component-based prompt architecture