Press → or Space to advance | ? for help
Welcome the audience. Introduce UALS as a next-generation adaptive learning platform built entirely on LLM technology. Emphasize that this is production-ready software, not a research prototype.
Key points: 100 intelligent agents, 45 total methods (15 KE + 15 SPL + 15 SATA), 3,375 unique learning combinations, 17 supported languages.
Duration: 1-2 minutes
🔗 All documentation available at /documentations/
Walk through the agenda. Mention that timing is flexible - 30 min for overview, up to 60 min with deep dives.
A Framework for AI-Enhanced Education
This analogy helps stakeholders understand why AI in education requires the same rigor as AI in medicine.
Supporting Multiple Educational Approaches
Domain → Subdomain → Concept
Same UI • Same Features • Same AI Support
Category → Competency → Proficiency Level
Critical architectural achievement: two fundamentally different philosophies on one platform without code duplication.
3,375 Unique Learning Combinations
Diverse ways to explore content: concept maps, problem-based scenarios, case studies, simulations...
Evidence-based pedagogies: Socratic dialogue, EMT feedback, cognitive apprenticeship...
Next-gen SATA formats: classic, priority ranking, weighted confidence...
This is NOT 3,375 pre-authored content paths – it's dynamically generated combinations.
Deep-dive exploration with dynamic concept mapping
Interactive tutoring with 5 agent types
Adaptive testing with IRT analytics
All three systems work in BOTH curriculum and competency philosophies
These three systems form the core learning experience. Click any card to open detailed documentation.
15 Methods for Content Exploration
Each method is grounded in cognitive load theory and multimedia learning principles.
15 Evidence-Based Pedagogical Approaches
Inquiry-based questioning
Expectation-Misconception-Tailored
4 Cs: Clarify, Question, Summarize, Predict
6 methods: Modeling to Exploration
Deep learning through elaboration
View all pedagogies
Effect sizes are from meta-analyses in educational research.
15 Next-Generation SATA Assessment Formats
Multiple correct answers
Order by importance
Certainty scoring
If-then reasoning
Timed responses
Multi-dimensional
No rounds - unlimited questions cycling through competencies
Individual metrics for each skill area
Based on most recent N questions
SATA formats reduce measurement error by 30-40% compared to traditional MC.
The Complete Adaptive Learning Ecosystem
🎯 Learning Insight Synthesis Agent (The 100th Agent) - Synthesizes data from all 99 monitoring agents
This is the most comprehensive ITS agent architecture in educational technology.
Domain Model, Curriculum Alignment, Misconception Detection
3 agentsKnowledge Tracing (BKT/PFA/DKT), Mastery Estimation, Affective State
4 agentsSocratic, EMT, Reciprocal Teaching, Cognitive Apprenticeship
9 agentsDialogue, Feedback, Hints, Scaffolding
5 agentsProblems, Worked Examples, Explanations
4 agentsPerformance Analytics, xAPI Tracking
2 agentsWalk through the 6 main categories. Emphasize Knowledge Tracing uses BKT, PFA, or DKT.
Production Software That Powers UALS
These agents work behind the scenes to ensure the 100 pedagogical agents can focus on learning.
"Show AI Thinking" - Real-time Decision Transparency
The Workflow Visualizer shows exactly which agents were invoked and what decisions were made.
5 Layers of Human-Verified Learning
Admin approval workflows, audit trails
Subject matter expert verification
Version control, edit tracking, approval
Reviewed content served, not raw LLM output
Prompt engineering, output validation
Students NEVER receive raw, unreviewed LLM output.
From Framework to Classroom
Select from school DSC catalog or create custom
Create personal copy for customization
Customize content generation instructions
Assign framework, set context
Share class link or code
Key architectural principle: LRS = Pointer (dscId), GCS = Data (actual competencies).
Every Interaction Tracked for Insights
/api/xapi-analytics/competency-performance/api/xapi-analytics/sata-analysis/api/xapi-analytics/report-card/api/xapi-analytics/learning-history/api/xapi-analytics/class-analyticsCRITICAL: Never fetch all statements (could be millions). Always use targeted queries.
Universal Templates • Dynamic DOM • Responsive UI
Express.js • Philosophy Detection • Route Handlers
100 ITS Agents • LLM Gateway • Orchestrator Engine
Google Cloud Storage • xAPI LRS • Cookieless Auth
Key principles: No local database, stateless for Cloud Run, cookieless auth for GDPR.
17 Languages • RTL Support • App-Wide
Usage: ?lang=zh or ?lng=es
RTL Languages: Arabic, Hebrew, Persian, Urdu automatically get dir="rtl"
LLM Integration: Language instruction injected into ALL prompts
When ?lang is set, the ENTIRE app must be in that language - no exceptions.
☁️ Stateless architecture enables unlimited horizontal scaling on Cloud Run
Token efficiency achieved through intelligent caching. Same competency + same config = same cache key.
Built on Decades of Learning Sciences
Every feature in UALS is grounded in peer-reviewed research.
All links open in new windows. Feel free to explore during Q&A.
Emphasize UALS is not just another chatbot - it's a complete intelligent tutoring system.
+10 agents: Transfer Learning, Memory Consolidation, Multimodal Orchestration, Essay Scoring, Code Review...
+10 agents: Learning Style, Cognitive Load, AR/VR, Voice Interaction, UDL...
+7 agents: Attention Management, Game-Based Learning, Haptic Feedback, Stealth Assessment...
Additional philosophies (Montessori, constructivist), domain-specific agents, LTI/SIS integration, research collaborations
The architecture supports N philosophies and unlimited agent additions.
Thank you for your attention!
UALS - Universal Adaptive Learning System
Open for questions. Common topics: gaming detection, content accuracy, implementation timeline, cost, data privacy.
This analogy establishes why AI in education requires the same rigor as AI in medicine:
| Medical Practice | UALS Education |
|---|---|
| Patients receive diagnosis | Learner model assessment via 100 agents |
| Prescription based on symptoms | Content recommendation based on knowledge gaps |
| Side effects monitored | Cognitive overload, frustration detected by affective agents |
| Treatment adjusted over time | Real-time pedagogical adaptation |
| Evidence-based medicine | Evidence-based pedagogy with effect sizes +0.30 to +0.90 |