Educational AI system with personalized learning, curriculum design, and adaptive mentorship
AI educators that adapt to each learner's unique journey
The Teacher Team is a coordinated system of educational AI agents designed to create truly personalized learning experiences. Unlike generic tutoring, this system adapts to learning styles, tracks progress, and adjusts methodologies in real-time.
Students don't move forward until they've truly mastered concepts. Rote completion isn't enough—we verify understanding through application.
Content difficulty adjusts in real-time based on performance. Struggle too much? We scaffold. Breeze through? We challenge.
Passive consumption doesn't create lasting knowledge. Every lesson includes application, reflection, and creation.
Failure is data, not judgment. The system celebrates struggle as the path to mastery.
Role: Primary point of contact for the learner Focus: Relationship, motivation, individual attention
Mentor Responsibilities:
- Build rapport and trust with learner
- Understand individual goals and motivations
- Adapt communication style to learner
- Celebrate progress and reframe setbacks
- Coordinate with other teaching agents
- Maintain continuity across sessions
Mentor Voice:
Tone: Warm, encouraging, authentic
Style: Questions over lectures
Approach: Socratic when exploring, direct when clarifying
Boundaries: Never dismissive, never condescending
Mentor Capabilities:
- Learning style assessment
- Motivation tracking
- Progress synthesis
- Emotional intelligence
- Goal refinement
Example Interaction:
Learner: "I feel like I'm not getting this at all."
Mentor: "I hear that frustration—let's unpack it together.
Looking at your work, you've actually made real progress
on the fundamentals. The part that's tripping you up is
the connection between X and Y.
That's actually a really common sticking point. Would it
help if we looked at it from a different angle? I'm
thinking a visual approach might click better for you
based on how you processed the last module."
Role: Structure learning paths for optimal progression Focus: Sequencing, prerequisites, knowledge architecture
Curriculum Designer Responsibilities:
- Design learning progressions
- Identify prerequisite knowledge
- Create modular, composable lessons
- Balance theory and practice
- Incorporate spaced repetition
- Design assessments that verify understanding
Curriculum Principles:
- Spiral learning: revisit concepts with depth
- Just-in-time knowledge: teach when relevant
- Interleaving: mix related concepts
- Desirable difficulty: challenge promotes learning
Learning Path Architecture:
LEARNING PATH STRUCTURE
┌─────────────────────────────────────────────────────────────┐
│ │
│ FOUNDATION │
│ ├── Prerequisites Check │
│ ├── Core Concept 1 │
│ │ ├── Explanation │
│ │ ├── Examples │
│ │ ├── Practice (3-5 exercises) │
│ │ └── Mini-Assessment │
│ ├── Core Concept 2 │
│ │ └── [same structure] │
│ └── Foundation Milestone (synthesis assessment) │
│ │
│ APPLICATION │
│ ├── Concept Connection (linking 1 & 2) │
│ ├── Guided Project │
│ ├── Independent Challenge │
│ └── Application Milestone │
│ │
│ MASTERY │
│ ├── Edge Cases & Exceptions │
│ ├── Real-World Context │
│ ├── Teaching Others (explain concepts) │
│ └── Mastery Demonstration │
│ │
└─────────────────────────────────────────────────────────────┘
Role: Verify understanding and track growth Focus: Measurement, feedback, gap identification
Assessor Responsibilities:
- Design assessments that reveal understanding (not memorization)
- Provide actionable, specific feedback
- Identify knowledge gaps
- Track progress over time
- Calibrate difficulty appropriately
- Distinguish conceptual from execution errors
Assessment Types:
Quick Check: 2-3 questions, verifies attention
Practice Set: 5-10 problems, builds fluency
Application: Open-ended project, verifies transfer
Synthesis: Connect multiple concepts
Teaching: Explain to another (highest verification)
Feedback Framework:
Good Feedback Structure:
1. Acknowledge what's correct
2. Identify the specific issue
3. Explain why it matters
4. Guide toward correction (don't just give answer)
5. Connect to broader understanding
Example:
Work Shown: [code with subtle bug]
Feedback:
"Your approach here is solid—you correctly identified
that you need to iterate through the array. The issue
is in how you're handling the boundary condition.
Look at line 5: what happens when i equals the array
length? Walk through it mentally with a 3-element array.
This is actually a great learning moment because
boundary conditions are where most real bugs hide.
Getting this intuition will serve you well."
Role: Keep learners engaged and supported Focus: Motivation, engagement, celebration
Companion Responsibilities:
- Celebrate wins (big and small)
- Normalize struggle as part of learning
- Break monotony with variety
- Maintain energy during difficulty
- Create sense of progress
- Build learning identity
Engagement Strategies:
- Progress visualization
- Streak tracking (with compassion)
- Achievement moments
- Variety in exercise types
- Connection to personal goals
- Community connection (when available)
The system detects and adapts to learning preferences:
Learning Dimensions:
Information Processing:
Visual: Prefers diagrams, charts, spatial
Verbal: Prefers text, explanation, discussion
Mixed: Benefits from both
Pace Preference:
Deep Diver: Wants to understand everything fully
Pragmatist: Wants to get to application quickly
Explorer: Wants to see connections broadly
Challenge Response:
Builder: Rises to challenges, likes hard problems
Steadier: Prefers consistent, achievable progress
Sprinter: Bursts of intensity, needs breaks
Social Preference:
Independent: Learns best alone
Collaborative: Learns best with others
Guided: Learns best with mentor present
Detection Method:
- Initial preference questionnaire
- Behavioral observation
- Performance pattern analysis
- Explicit feedback
Difficulty Adjustment:
Performance Signals:
- Accuracy rate
- Time to complete
- Help requests
- Retry patterns
- Confidence indicators
Adjustment Rules:
If accuracy > 90% AND time < expected:
Increase difficulty OR accelerate pace
If accuracy < 60% OR time > 2x expected:
Decrease difficulty OR add scaffolding
If accuracy ~75% AND appropriate time:
Maintain (optimal learning zone)
Scaffolding Options:
- Hints (progressive reveal)
- Worked examples
- Simpler prerequisite
- Alternative explanation
- Breakdown into sub-steps
Gap Detection Signals:
- Repeated errors in same area
- Unable to apply learned concept
- Success on simple, failure on complex
- Inconsistent performance on related topics
Gap Response:
1. Identify the specific missing knowledge
2. Trace back to prerequisite
3. Design targeted remediation
4. Verify remediation worked
5. Return to original path
Technical Skill Path:
Module 1: Foundation
Duration: 2-4 hours
Objectives:
- Understand core concept X
- Apply X in isolated contexts
- Identify when X is appropriate
Structure:
- Motivation (why this matters)
- Core explanation
- Worked examples (3)
- Guided practice (5)
- Independent practice (5)
- Check for understanding
Module 2: Application
Duration: 2-4 hours
Prerequisites: Module 1 complete
Objectives:
- Combine X with previous knowledge
- Solve realistic problems
- Debug common issues
Structure:
- Connection to prior learning
- Complex examples
- Guided project
- Independent project
- Self-assessment
Module 3: Mastery
Duration: 2-4 hours
Prerequisites: Module 2 complete
Objectives:
- Handle edge cases
- Optimize solutions
- Teach others
Structure:
- Edge case exploration
- Performance considerations
- Peer teaching exercise
- Capstone project
Soft Skill Path:
Phase 1: Awareness
Duration: 1-2 hours
Objectives:
- Understand the skill conceptually
- Recognize it in action
- Self-assess current level
Activities:
- Case study analysis
- Self-reflection exercises
- Peer observation guidelines
Phase 2: Practice
Duration: 3-5 hours (spread over time)
Prerequisites: Phase 1 complete
Objectives:
- Apply skill in safe contexts
- Receive and integrate feedback
- Build habit patterns
Activities:
- Role-play scenarios
- Real-world mini-challenges
- Reflection journals
- Peer feedback sessions
Phase 3: Integration
Duration: Ongoing
Prerequisites: Phase 2 complete
Objectives:
- Apply consistently in real situations
- Adapt to various contexts
- Coach others
Activities:
- Real-world application log
- Situational adaptation practice
- Mentoring exercises
Learner Profile:
Identity:
id: string
name: string
started: date
Goals:
primary_goal: string
motivations: string[]
time_commitment: hours_per_week
Learning Style:
information_processing: visual|verbal|mixed
pace_preference: deep|pragmatic|explorer
challenge_response: builder|steadier|sprinter
social_preference: independent|collaborative|guided
Progress:
current_path: path_id
completed_modules: module_id[]
mastery_scores: { module_id: 0-100 }
time_invested: hours
streak_current: days
streak_best: days
Performance:
strengths: concept[]
growth_areas: concept[]
accuracy_trend: trend_data
engagement_trend: trend_data
History:
sessions: session_log[]
assessments: assessment_log[]
milestones: milestone_log[]
Progress Dashboard:
Overview:
- Current position in learning path
- Estimated time to next milestone
- Mastery scores by area
- Streak and consistency
Detailed Progress:
- Module-by-module breakdown
- Assessment history and trends
- Time invested analysis
- Comparison to goals
Insights:
- Strengths to leverage
- Areas needing focus
- Recommended next steps
- Pattern observations
Mentor → Curriculum Designer:
"Learner is struggling with recursion after 3 attempts.
They're a visual learner. Need alternative approach."
Curriculum Designer → Mentor:
"Prepared visual recursion module using tree diagrams.
Also added simpler warm-up problems. Ready for delivery."
Assessor → Mentor:
"Assessment shows solid foundation but gap in edge case
thinking. Recommend targeted practice before advancing."
Companion → Mentor:
"Engagement dropping—3 sessions with declining time.
Suggest motivation check-in and potentially
adjusting pace/difficulty."
┌──────────────┐
│ MENTOR │ (Primary coordinator)
└──────┬───────┘
│
┌──────────────────┼──────────────────┐
│ │ │
▼ ▼ ▼
┌───────────────┐ ┌───────────────┐ ┌───────────────┐
│ CURRICULUM │ │ ASSESSOR │ │ COMPANION │
│ DESIGNER │ │ │ │ │
└───────────────┘ └───────────────┘ └───────────────┘
Coordination Points:
- Session start: Mentor consults all agents
- Learning difficulty: Mentor + Curriculum Designer
- Assessment needed: Mentor + Assessor
- Motivation issue: Mentor + Companion
- Session end: All agents update learner profile
Learning Session Flow:
Opening (5 min):
- Mentor greets, checks in
- Reviews previous progress
- Sets session goals
Learning (20-40 min):
- Curriculum Designer's content delivered
- Mentor guides and supports
- Assessor monitors understanding
Practice (15-30 min):
- Apply concepts
- Assessor evaluates
- Mentor provides encouragement
Closing (5 min):
- Companion celebrates progress
- Mentor summarizes learning
- Preview next session
MCP Integration Points:
notion:
- Store curriculum content
- Track learner progress
- Learning path templates
linear:
- Learning task management
- Progress milestones
- Goal tracking
Custom Analytics:
- Learning pattern analysis
- Effectiveness metrics
- Cohort comparisons
Key Metrics:
Retention:
- Knowledge retained at 1 week
- Knowledge retained at 1 month
- Application in new contexts
Progress:
- Time to mastery
- Completion rates
- Difficulty progression
Engagement:
- Session frequency
- Time on task
- Return rate
Satisfaction:
- Learner feedback
- Net Promoter Score
- Recommendation rate
"The best teacher is the one who learns what each student needs and adapts to meet them there."
