Case study / ISTE+ASCD
Build the learning.
Improve the way
the work gets done.
I developed 100+ learning experiences in six months while creating AI-assisted workflows that colleagues across my department adopted.
01 / The challenge
A high-volume learning portfolio. A short production window.
Producing learning at scale takes more than generating content. Source material has to become a coherent experience, with consistent structure, meaningful activities, functional interactions, and clear review steps.
I was responsible for developing learning products under tight deadlines. I built workflows to make recurring production and review tasks more repeatable while retaining my role in design, implementation, and QA.
02 / My ownership
From source content to a review-ready experience.
What I owned
- Course architecture and consistent page designs.
- D2L Brightspace builds and interactive activities.
- AI agent instructions and production workflows.
- Accessibility checks, internal QA, and revisions.
- Demonstrations and individual coaching for colleagues.
How the handoff worked
My director used an AI agent to prepare source content. I used the source material and my workflows to design and build the learning experience.
My director and managing director reviewed the work before it went live. Production turnaround refers to my build and internal QA before that final review.
03 / The system
Five connected workflows.
Clear roles for AI and people.
Prepare the source
Review source documents for grammar, terminology, structural alignment, assessment quality, and consistency before production.
Build the course
Translate prepared material into D2L-ready pages using reusable page patterns, role pathways, and platform compatibility rules.
Design immersive practice
Develop distinct demonstrations, guided practice, and independent application. Define learner actions, feedback, and progression, rather than adding interactions for decoration.
Revise with control
Identify page-level changes across accessibility, copy, functionality, and design. Review a change list before applying revisions.
Check quality and accessibility
Structure the review around evidence, reviewer feedback, and remediation priorities. Separate identified issues from items requiring manual verification.
AI support does not replace verification.
The workflow includes human review and explicit limits on what has been verified. An automated check does not, by itself, establish accessibility conformance or content accuracy.
04 / Production results
A shorter path to review-ready learning.
| Production scope | Estimated time |
|---|---|
| Before AI agent workflows | Approximately one month to develop a course. |
| A 2–5-hour course with the workflows | Approximately one workday, including design, build, accessibility checks, and my QA. |
| A 15-hour course with the workflows | Approximately two workdays for the same production scope. |
These are my reported estimates, not a controlled comparison of equivalent courses. Course hours describe estimated learner duration. Production times exclude director and managing director review and approval.
Across six months, I developed more than 100 learning experiences spanning courses, Strategies, and Bites. The workflows supported faster production, more consistent design, and a more structured approach to accessibility review.
05 / Adoption & leadership
The work extended beyond my own desk.
Colleagues across my department adopted the workflows. I supported that adoption through demonstrations and individual coaching, helping people understand how to use the process in their own work.
This is where learning design and AI adoption meet in my practice: turning a tool into a usable workflow, explaining the decisions behind it, and helping people apply it with human judgment.
06 / Explore the approach
Try an original learning experience.
This case study describes my role and process without reproducing internal course materials or agent instructions. For a hands-on example, explore a separate portfolio simulation with original fictional content.
In the Participation Control Room, you're a colleague helping a quiet team discuss AI concerns and agree on a limited brainstorming experiment. The experience demonstrates contextual practice, branching feedback, and replay.
Play the Participation Control Room