Portfolio Examples Generative AI
Case Study · AI-Assisted Learning Design

Generative AI.

Two learning experiences exploring practical and innovative applications of generative AI in instructional design.

AI-Assisted Development · Scenario-Based Learning · Interactive Design · Cross-Functional Collaboration

Project Overview

These projects explore how generative AI can support authentic decision-making practice, application of course concepts, and formative feedback.

I collaborated with faculty subject matter experts (SMEs) and instructional design colleagues to translate learning objectives into interactive, scenario-based learning experiences using emerging AI-assisted development tools.

The work involved iterative prototyping with unfamiliar technology while considering scalability, faculty workload, and implementation constraints across different enrollment levels.

The Instructional Challenges

Both projects address a common learning design challenge: providing realistic decision-making practice and timely formative feedback while accounting for course requirements, faculty workload, and scalability.

In MGT 411: Leading Organizations, students needed opportunities to recognize cognitive biases and consider how those biases influence managerial decisions—without having mistakes affect their grades.

In WPC 480, a high-enrollment capstone course, we explored how learners could practice navigating difficult leadership situations through ongoing conversations, with the flexibility to use the experience for low-stakes practice or instructor-evaluated work.

Design Process / Canvas + Codex

My Workflow

This seven-step workflow moves from an observable learning objective to an interactive Canvas activity. I define the instructional requirements and approve the activity brief; Codex supports prototyping and iteration; and I review accessibility, formative feedback, and LMS tracking before release.

From learning objective to interactive — Canvas and Codex workflow A human-led seven-step collaboration: 1. Set the learning target: Name one observable learner action tied to an objective. 2. Choose the learner task: Give learners a realistic decision to make. 3. Approve the activity brief: Agree on directions, feedback, scoring, and flow. 4. Build the full experience: Create a working path from introduction to completion. 5. Review, revise, repeat: Use screenshots and focused learner-facing feedback. 6. Check access and tracking: Test keyboard use, feedback, and Canvas reporting. 7. Test in Canvas and release: Verify launch, completion, and gradebook results. Course designer sets the purpose, judges accuracy, and approves release. Codex suggests, drafts, builds, checks, and revises. Source: Building Accessible Canvas Interactives With Codex. DESIGN PROCESS / CANVAS + CODEX From learning objective to interactive A human-led, seven-step collaboration 1 Set the learning target Name one observable learner action tied to an objective. 2 Choose the learner task Give learners a realistic decision to make. 3 Approve the activity brief Agree on directions, feedback, scoring, and flow. 4 Build the full experience Create a working path from introduction to completion. 5 Review, revise, repeat Use screenshots and focused learner-facing feedback. 6 Check access + tracking Test keyboard use, feedback, and Canvas reporting. 7 Test in Canvas + release Verify launch, completion, and gradebook results. THE PARTNERSHIP Course designer: sets the purpose, judges accuracy, approves release. Codex: suggests, drafts, builds, checks, and revises. Source: Building Accessible Canvas Interactives With Codex
DESIGN PROCESS / CANVAS + CODEX

From learning objective to interactive

A human-led, seven-step collaboration

  1. 1
    Set the learning target

    Name one observable learner action tied to an objective.

  2. 2
    Choose the learner task

    Give learners a realistic decision to make.

  3. 3
    Approve the activity brief

    Agree on directions, feedback, scoring, and flow.

  4. 4
    Build the full experience

    Create a working path from introduction to completion.

  5. 5
    Review, revise, repeat

    Use screenshots and focused learner-facing feedback.

  6. 6
    Check access + tracking

    Test keyboard use, feedback, and Canvas reporting.

  7. 7
    Test in Canvas + release

    Verify launch, completion, and gradebook results.

THE PARTNERSHIP

Course designer: sets the purpose, judges accuracy, approves release.

Codex: suggests, drafts, builds, checks, and revises.

Source: Building Accessible Canvas Interactives With Codex

This workflow illustrates my human-led approach to AI-assisted development: instructional decisions, quality checks, and release approval remain with the course designer.

Bias Analysis Micro-Case Widget

An ungraded, scenario-based learning activity for MGT 411 that supports decision-making practice, multiple response paths, and formative feedback.

Instructional Design

I helped develop an interactive micro-case in which learners analyze a management scenario, identify potential cognitive biases, and select a response. Formative feedback and a debrief connect their decisions to course concepts.

The activity provides a low-stakes practice environment where learners can explore multiple scenarios and response options without affecting their course grades.

Implementation & Evaluation

Developed in early 2026, the widget is currently in implementation in MGT 411. A mixed-methods evaluation is planned, combining qualitative course analysis with quantitative learner feedback to be collected through Qualtrics.

Optional learner transcript submissions are under consideration as a source of qualitative, formative feedback on decision-making. This feature has not been implemented.

Evaluation findings and learner performance outcomes are not yet available.

View Bias Analysis Project

Boss Level Leadership Simulator

An AI-supported, scenario-based leadership simulation with open-ended dialogue, competing priorities, and unexpected workplace situations.

Instructional Design

The learner assumes the role of a newly hired manager and is quickly contacted by an employee. Through ongoing conversations, they get to know a team with different personalities, needs, and performance challenges—including an employee dealing with mental illness, someone who has become disengaged, someone trying to improve poor performance, and someone creating conflict with coworkers.

The simulation can also introduce unexpected events, including urgent HR matters or impromptu all-hands meetings. Learners make contextual decisions through open-ended dialogue rather than fixed multiple-choice branches.

Assessment & Feedback

The working prototype generates an end-of-day performance review. Its current feedback is broad and scale-based, rather than tied to a detailed assessment rubric.

The next iteration is planned to provide more specific, rubric-informed formative feedback based on each learner’s decisions, with responses remaining in character to preserve immersion.

Potential delivery models include low-stakes practice or a faculty-evaluated assignment using conversation transcripts. Any assessment model would depend on course objectives, enrollment, and available evaluation capacity.

Current Status

Boss Level remains a working prototype. Open-ended dialogue and the end-of-day review are functional; a refined assessment rubric and more context-sensitive feedback are planned enhancements.

View Boss Level Project

Development & Collaboration

I collaborated with faculty subject matter experts (SMEs) to establish content requirements and align the activities with course learning objectives.

Working with my director at the time, I explored Codex as an AI-assisted development tool. We used iterative prototyping and testing to translate instructional concepts into functional interactions while identifying technical capabilities and constraints.

My contribution focused on aligning technical design choices with learning objectives, collaborating with SMEs, and iteratively refining the learner experience.

Outcomes & Next Steps

The projects are at different delivery stages. Bias Analysis is in course implementation, with a planned mixed-methods evaluation to inform future iterations. No validated learner performance outcomes are available yet.

Boss Level is a functioning prototype. The next development priorities are a refined assessment rubric and more contextual, in-character feedback that supports learner reflection.

Together, the projects illustrate two approaches to practice-based learning: structured decision-making micro-cases and an open-ended, conversational leadership simulation.

Project Takeaways

These projects reinforced the importance of starting with the learning need, then assessing emerging technology against instructional objectives, accessibility considerations, and delivery constraints. Iterative prototyping and SME collaboration helped translate design concepts into interactive practice experiences.

My focus is on AI-supported learning that provides meaningful practice and formative feedback while remaining practical to implement, facilitate, and evaluate.