Learner UX Architecture

My Role

• Complete learner experience architecture across 8+ distinct workflow areas
• Technical advisor bridging UX and engineering teams
• System analysis, user flows, wireframes, visual design, and prototyping
• Architecture documentation connecting learner experience to the broader ecosystem
• Stakeholder presentations to initiative leadership
• Co-inventor on patent application (USPTO 18/313268)

Results

The Challenge

Computer-based training and web-based learning have existed for decades, but most systems are fundamentally flawed. They are not much more than elaborate PowerPoint presentations: a pre-scripted path from A to Z. If you already know the concept and want to skip to the knowledge check, you cannot. If life gets in the way and you need to pause, you lose your progress. If your goal changes mid-program, the system does not care.

We wanted to build something genuinely better. A learner experience that recognized what students already knew, adapted to their goals, and flexed around their lives instead of forcing them into rigid schedules and predetermined sequences.

The Goal

Design a complete learner experience that treated education as personal, adaptive, and goal-driven. Every aspect of the learner’s journey, from discovering programs to earning credentials, needed to work as a unified experience powered by machine learning and rich metadata. Not a better LMS, but a fundamentally different relationship between learner and institution.

The Constraints

This was a greenfield project with no existing foundation to build from. The internal team was new, with no established processes, no deployment infrastructure, and no QA team. Contract developers worked part-time. And the learner experience had to integrate seamlessly with the broader ecosystem being built simultaneously by other teams.

Our Users

We operated under the belief that everyone should be a lifelong learner. That philosophy shaped even the language we used: we referred to “students” as “learners” throughout the system.

Traditional vs. Our Approach

Every major LMS platform (Blackboard, Canvas, Moodle, Cornerstone, Workday Learning) shares the same foundational architecture: static courses, linear pathways, passive consumption. We challenged every assumption.

From Napkin Sketch to Architecture

Early brainstorming captured the key pieces and flows. From those initial concepts I built a comprehensive Learner UX Architecture: every touchpoint, every workflow, every system interaction across the complete learner journey. These diagrams became the guide for the teams building enrollment, learning, assessment, and credentialing, showing each team exactly how their piece connected to the whole.

The Solutions

This page focuses on the learner experience: the five workflows that make up a student’s journey from discovery through credentialing. For the broader context (organization, competitive landscape, AI philosophy, full ecosystem architecture), see the AI-Powered Educational Ecosystem case study.

Rather than patching a failing model, we designed a fundamentally different approach to education. AI would power personalization and streamline administrative processes (not generate content, but guide students through complexity). Credit for Prior Learning would recognize what students already knew, eliminating redundant coursework. The model shifted from degree-oriented to goal-oriented: students pursuing specific competencies and outcomes, not arbitrary credit hours. And we removed the rigid term structure that forced students to choose between life and education, replacing it with flexibility that adapted to their reality.

1. Public Site and Guide Me

Before enrollment, learners access the university through marketing content and a program catalog. The centerpiece was the Guide Me Experience: an AI-powered onboarding workflow where learners enter as much or as little as they want, including past experience, education, career aspirations, personal preferences, and even hobbies and interests.

The system responds with multiple pathway options, showing expected income per path, career opportunities, time to goal, and alternative routes. The more data a learner provides, the more personalized the recommendations. This was not a simple quiz that slots you into a program. It was a conversation between the learner’s goals and the system’s understanding of what pathways could get them there.

2. Learner Home

The central hub for managing the entire university experience. Every tool a learner needed, in one place.

3. Learning Pathway

This is where the system fundamentally diverges from traditional LMS platforms. The learning pathway uses multiple data sources to build and continuously manage each learner’s unique sequence: Guide Me responses, Credit for Prior Learning records, assessment results, progress data, and changed goals.

Successes complete steps. Failures provide supplemental materials rather than dead ends. Nudges keep learners motivated on their own schedules. And the pathway recalculates on any change, always showing the most efficient route to the learner’s goal. A dropped course does not just disappear; the system models the downstream impact and suggests alternatives. Counselors and learners both get the visibility they never had before.

4. Credit for Prior Learning

Machine learning (ML) algorithms inspect transcripts and references, processing them to create CPL records automatically. Learners upload transcripts, certifications, and resumes. The ML engine maps them to competencies. Registrars verify rather than manually process from scratch. The pathway adjusts: modules are bypassed or supplemental content is suggested.

Critically, new CPL can be added anytime, not just at enrollment. As a learner gains experience mid-program, the system can recognize it. The CPL engine also addressed a real operational constraint: the university’s registrar staff was at capacity. Automating the initial evaluation step meant scaling enrollment without scaling headcount.

5. Goal Matching and Achievements

The system proactively suggests optimizations as learners progress: additional credit that opens better career matches, an extra course that unlocks higher pay, earlier workforce entry options, or alternative pathways better suited to evolving goals. The pathway never stopped working on the learner’s behalf.

As learners progressed, they earned achievements: on-screen moments of recognition, certificates, course credits, and degrees. Achievements accumulated and stacked toward larger credentials, making every step of progress visible and meaningful. Progress was measured through demonstrated competency, not seat time or attendance.

An Intelligent Approach to AI

While competitors rushed to use AI for content generation (creating entire courses, tests, and learning materials), we saw the risk. AI hallucinations in educational content are not just problematic; they are dangerous.

We took a different approach entirely: do not use AI to create content. Instead, apply machine learning to administrative and guidance systems where it could genuinely help students navigate their education. Our models processed vast amounts of transcript data, enrollment records, and competency frameworks, with human oversight at every phase to catch errors before they impacted students.

Three specific applications: accelerating transcript evaluation from weeks to days, suggesting personalized learning paths based on each student’s background and goals, and intelligently mapping coursework to the competencies they needed. The AI handled the complexity of matching prior learning to requirements, not the learning itself.

Approach Decisions

Beyond iterating on individual designs, we made strategic decisions about system architecture. One example: how to model associations between students and staff. I identified both an ideal approach and a simpler, faster one. For the first iteration, our priority was standing the system up and proving it worked, so we chose the expedient path. It was less robust, but it solved the core problem and got us to validation faster.

As students complete their learning pathway, different courses belong to different disciplines, and each discipline has a different instructor. The expedient choice left the system inflexible to those real-world associations. Documenting that tradeoff explicitly, and designing the ideal model alongside it, meant the engineering team knew exactly what to build next when resources allowed.

Final Designs

Working with talented design teams, multiple groups tackled specific experience areas while I continued architecture work on the overall ecosystem, the authoring platform, and the communication center.

What We Proved

• POC: Core learner flows validated end-to-end, from enrollment through pathway navigation to assessment
• Alpha Release: Internal users completed full learner journeys through the system
• Architecture Validated: Learner experience integrated successfully with ecosystem components
• Level Up Labs Pilot: Two-thirds of pilot course content completed using general education modules to exercise end-to-end functionality

What was planned before the department closed: real user testing with actual learners, expansion beyond pilot content, and CPL at scale with real student transcripts processed through the ML engine.

Patent

I was a primary contributor to the SNHU patent application:

“Technologies and services to deliver customized and responsive learning pathways and related systems and methods”

Co-inventors: Joel Cory, Michael Moretti (Technical PM), Lisa Hodge (UX Writer)

USPTO Patent Application 18/313268

Reflections

This was truly personalized learning. Not just adaptive content delivery, but a complete reimagining of how education could work. Every component was designed to serve the individual learner while scaling to 135,000+ students.

Architecting AI-integrated user experiences requires a different kind of thinking than traditional UX. The system’s behavior is not static; it changes based on every learner’s unique data. Designing for that means thinking in rules and relationships, not just screens and flows. Balancing personalization with scalability was a constant tension: every feature that made the experience more personal also made the system more complex. The metadata architecture was the key to resolving that tension, letting us create flexibility at the system level without requiring custom design for every learner scenario.

This project also taught me how to design for multiple user types within deeply interconnected systems. The learner experience could not be designed in isolation; every decision rippled into advisor workflows, assessment processes, and authoring requirements.

SNHU ultimately closed large portions of its business due to financial pressures. The irony was not lost on us: the very problems driving the university’s difficulties (declining enrollment, poor retention, manual processes that could not scale) were exactly what we were building solutions for.

Related Projects

• AI-Powered Educational Ecosystem – Overall ecosystem architecture
• Authoring Platform – Content creation system
• Achievement Process – Credentialing system