AI Image Generator

Designing predictive AI to reduce cognitive load for teachers

What?

AI image generator that creates classroom-ready visuals by anticipating what you need next.

For who?

K-12 and higher-education teachers preparing slides, worksheets, and LMK posts under tight prep time.

My role

Product designer owning research, IA, flows, and UI

Project duration

5 weeks

Tools

Figma

Why This Matters

Getting tailored results from existing AI models is not as easy as one might think, and it's costing teachers important time that could be spent with students.

Most teachers are not AI prompt experts which can make it difficult to get specific results, especially when AI generators are making assumptions to fill in the gaps. Generic AI tools are not tuned for grade level, curriculum, or safety, which slows lesson prep and leads to inconsistent visuals.

What Educators Need

Produce age-appropriate, on-topic images quickly.

Produce age-appropriate, on-topic images quickly.

Avoid unsafe or copyright-sensitive content.

Avoid unsafe or copyright-sensitive content.

Keep a consistent style across slides without prompt gymnastics.

Keep a consistent style across slides without prompt gymnastics.

The Solution

An anticipatory image generator that turns Subject, Grade, and Learning Goal into smart suggestions so teachers get the right image in fewer prompts.

An anticipatory image generator that turns Subject, Grade, and Learning Goal into smart suggestions so teachers get the right image in fewer prompts.

How I Did It

Establish Goals

Empower educators

Enable teachers, instructional designers, and educators to generate custom visuals without needing advanced technical skills.

Save time

Replace time-consuming manual searches or edits with a fast, AI-driven process so that educators can spend time engaging with their students.

Ensure tailored results

Provide outputs that are appropriate for classroom use, incorporating robust content filters and educator-specific style presets.

Competitive Analysis

I analyzed four existing AI image tools (Canva, Freepik, Night Café, and Google Gemini) to understand how educators might currently approach AI-generated visuals. While these tools offer strengths such as templates, stock libraries, and artistic outputs, none address the specific needs of teachers.

The gaps in these models highlighted an opportunity to design an image generator that combines speed, classroom-appropriate assets, and anticipatory suggestions, helping teachers quickly reach a usable visual with less trial and error.

I analyzed four existing AI image tools (Canva, Freepik, Night Café, and Google Gemini) to understand how educators might currently approach AI-generated visuals. While these tools offer strengths such as templates, stock libraries, and artistic outputs, none address the specific needs of teachers.

The gaps in these models highlighted an opportunity to design an image generator that combines speed, classroom-appropriate assets, and anticipatory suggestions, helping teachers quickly reach a usable visual with less trial and error.

Iteration 1

Iteration 1

Simplifying the Input Flow

Simplifying the Input Flow

Simplifying the Input Flow

Problem: Existing flows ask teachers for too many inputs, creating friction.

Change: Reduced setup to three essentials: Grade, Subject, Style.

Impact: Teachers could set up a tailored request in under 30 seconds before starting the text-based input, a faster start compared to existing AI tools.

Problem: Existing flows ask teachers for too many inputs, creating friction.

Change: Reduced setup to three essentials: Grade, Subject, Style.

Impact: Teachers could set up a tailored request in under 30 seconds before starting the text-based input, a faster start compared to existing AI tools.

Early Exploration

Wireframes

  • Step 1: User selects cards to indicate grade level and subject/learning goal.

  • Step 2: User selects a card to indicate the type of image they want to create.

  • Step 3: User chooses another card to indicate image style.

  • Step 4: User's previous selections are kept as tags at the top of the screen. They are then brought to a text-based input sequence to interact with the AI agent.

  • Step 1: User selects cards to indicate grade level and subject/learning goal.

  • Step 2: User selects a card to indicate the type of image they want to create.

  • Step 3: User chooses another card to indicate image style.

  • Step 4: User's previous selections are kept as tags at the top of the screen. They are then brought to a text-based input sequence to interact with the AI agent.

Feedback

Feedback

Educators liked the simple flow, but felt unsure how to refine results without starting over.

Next, we focused on adding guided suggestions to make refinement easier and reduce prompt entries.

Iteration 2

Iteration 2

Adding Anticipatory Suggestions

Adding Anticipatory Suggestions

Adding Anticipatory Suggestions

Problem: Teachers weren't sure how to refine images without starting over, which led to multiple trial-and-error prompts.

Change: Introduced one-tap suggestion chips to guide refinement without retyping.

Impact: In testing, educators reported greater confidence and reduced retries by 40%, reaching a usable image in fewer steps.

Early Exploration

Wireframes

  • Initial Inputs: Teachers start by entering grade level and subject, giving the AI a strong foundation for tailored results.

  • First round of suggestions: The AI suggests possible prompt refinements right away, helping teachers move forward faster without needed to rewrite prompts.

  • Style selection for consistency: Teachers choose a style once and apply it across materials, ensuring visual consistency throughout their lesson content.

  • Confirmation setup: Once a suggestion is accepted, the AI confirms choices to ensure accuracy before generating the image, reducing misaligned outputs.

  • Start of image generation: After inputs are confirmed the AI begins producing an image, giving teachers a clearer sense of progress in the flow.

  • Initial Inputs: Teachers start by entering grade level and subject, giving the AI a strong foundation for tailored results.

  • First round of suggestions: The AI suggests possible prompt refinements right away, helping teachers move forward faster without needed to rewrite prompts.

  • Expanded guidance: If the teachers reject a suggestion, the system provides additional options, preventing dead ends in the workflow.

  • Confirmation setup: Once a suggestion is accepted, the AI confirms choices to ensure accuracy before generating the image, reducing misaligned outputs.

  • Start of image generation: After inputs are confirmed the AI begins producing an image, giving teachers a clearer sense of progress in the flow.

Feedback

Feedback

Splitting the flow into multiple steps risked slowing users down, and the text input box was underused since inputs were already handled in the message boxes.

Next, I aimed to front-load key information and give the text box a clearer purpose to streamline that experience.

Final Design

Final Design

Faster, Classroom-Ready Image Generation

Faster, Classroom-Ready Image Generation

Faster, Classroom-Ready Image Generation

Problem: Previous flows required multiple steps and left educators uncertain how to refine or reuse inputs, slowing down lesson prep.

Change: Consolidating the setup into a single screen with front-loaded inputs (Grade, Subject, Topic/Learning Goal) and added anticipatory suggestion chips.

Impact: Educators could generate a classroom-ready image in fewer steps, with higher first-try success and less fatigue, making the process faster, more reliable, and tailored to lesson needs.

Wireframes

  • Front-loaded inputs: Teachers provide key information upfront, reducing friction later and setting the AI up for more accurate suggestions.

  • Smart image suggestions: The system anticipates needs by suggesting classroom-friendly image types (diagram, chart, illustration), cutting down trial-and-error prompt crafting.

  • Style selection for consistency: Teachers choose a style once and apply it across materials, ensuring visual consistency throughout their lesson content.

  • Flexible text input: The text box is reframed as an optional detail-attention tool, giving educators flexibility without requiring heavy typing.

  • First image preview + refinement options: AI generates an initial image and offers one-tap refinements, reducing retries and helping teachers get to a usable result faster.

  • Iterative refinements: Teachers can continue to refine and polish their images with guided options or finish generating, avoiding the fatigue of re-prompting from scratch.

  • Export + save for lesson use: Final images can be downloaded, added to a lesson plan, or stored in a library, directly supporting classroom workflows.

  • Front-loaded inputs: Teachers provide key information upfront, reducing friction later and setting the AI up for more accurate suggestions.

  • Smart image suggestions: The system anticipates needs by suggesting classroom-friendly image types (e.g., diagram, chart, illustration), cutting down trial-and-error prompt crafting.

  • Style selection for consistency: Teachers choose a style once and apply it across materials, ensuring visual consistency throughout their lesson content.

  • Flexible text input: The text box is reframed as an optional detail-attention tool, giving educators flexibility without requiring heavy typing.

  • First image preview + refinement options: AI generates an initial image and offers one-tap refinements, reducing retries and helping teachers get to a usable result faster.

  • Iterative refinements: Teachers can continue to refine and polish their images through guided options or move on to finalizing and downloading assets, avoiding the fatigue of re-prompting from scratch.

  • Export + save for lesson use: Final images can be downloaded, added to a lesson plan, or stored in a library, directly supporting classroom workflows.

Final Prototype

This prototype demonstrates how educators can go from entering lesson details to generating a classroom-ready image in just a few guided steps, with anticipatory suggestions reducing retries and ensuring consistency.

This prototype demonstrates how educators can go from entering lesson details to generating a classroom-ready image in just a few guided steps, with anticipatory suggestions reducing retries and ensuring consistency.

Reflection

Reflection

Reflection

Working with Reality AI was my first time working with AI models and conversational agents, which challenged me to think beyond traditional interfaces and explore how predictive design could reduce user effort. Designing AI agents taught me how to approach emerging technologies with curiosity while still grounding decisions in user needs.

Key takeaways from this project include:

  • Learning how to balance AI-driven guidance with user control to prevent fatigue

  • Structuring flows that anticipate intent while still offering flexibility

  • Framing complex technologies in a way that feels simple and approachable for non-technical users like educators

This experience expanded my design toolkit and gave me the confidence to tackle projects that integrate cutting-edge technology into everyday workflows. Moving forward, I'll continue to apply these lessons by making new technologies feel trustworthy, accessible, and intuitive for diverse user groups.

Working with Reality AI was my first time working with AI models and conversational agents, which challenged me to think beyond traditional interfaces and explore how predictive design could reduce user effort. Designing AI agents taught me how to approach emerging technologies with curiosity while still grounding decisions in user needs.

Key takeaways from this project include:

  • Learning how to balance AI-driven guidance with user control to prevent fatigue

  • Structuring flows that anticipate intent while still offering flexibility

  • Framing complex technologies in a way that feels simple and approachable for non-technical users like educators

This experience expanded my design toolkit and gave me the confidence to tackle projects that integrate cutting-edge technology into everyday workflows. Moving forward, I'll continue to apply these lessons by making new technologies feel trustworthy, accessible, and intuitive for diverse user groups.

Other Projects