Performance Tracking

Overview

The platform uses the UserMastery model to track your learning progress across different topics and concepts. This comprehensive tracking system helps you identify strengths, weaknesses, and areas requiring focused practice.

UserMastery Model

Each topic you practice has a corresponding UserMastery record:

class UserMastery(db.Model):
    # Identifiers
    user_id                 # Your user ID
    topic                   # Topic name (e.g., 'machine_learning', 'python')
    
    # Session tracking
    sessions_attempted      # Number of practice sessions
    questions_attempted     # Total questions answered
    correct_count           # Number of correct answers
    last_session_date       # Most recent practice date
    
    # Mastery scores (0-1 range)
    mastery_level          # Overall mastery score
    semantic_avg           # Average semantic similarity
    keyword_avg            # Average keyword coverage
    
    # Performance metrics
    avg_response_time      # Average time to answer
    
    # Learning velocity
    mastery_velocity       # Rate of improvement
    last_mastery           # Previous mastery score
    
    # Adaptive difficulty
    current_difficulty     # Current difficulty level
    consecutive_good       # Streak of good performances
    consecutive_poor       # Streak of poor performances
    
    # Concept-level tracking
    concept_masteries      # JSON with detailed concept data
    weak_concepts          # Concepts below mastery threshold
    strong_concepts        # Concepts above mastery threshold
    concept_stagnation     # Concepts not improving

Viewing Mastery Levels

Overall Progress

Get a comprehensive view of your progress across all topics:

GET /api/user/progress
Authorization: Bearer <your-token>

Response Structure:

{
  "topics": {
    "machine_learning": {
      "mastery": 0.785,
      "sessions_attempted": 12,
      "questions_attempted": 96,
      "correct_count": 78,
      "last_session_date": "2026-03-03T14:30:00",
      "semantic_avg": 0.802,
      "keyword_avg": 0.768,
      "weak_concepts": ["neural_networks", "gradient_descent"],
      "strong_concepts": ["linear_regression", "decision_trees"],
      "current_difficulty": "medium",
      "mastery_velocity": 0.042
    },
    "python": {
      "mastery": 0.892,
      "sessions_attempted": 8,
      "questions_attempted": 64,
      "correct_count": 59,
      "weak_concepts": ["decorators"],
      "strong_concepts": ["list_comprehensions", "generators", "classes"]
    }
  },
  "overall": {
    "avg_mastery": 0.839,
    "total_sessions": 20,
    "topics_count": 2
  }
}

Located in app.py:1992-2081

Topic-Specific Mastery

Get detailed mastery data for a specific topic:

GET /api/adaptive/mastery/<topic>
Authorization: Bearer <your-token>

Response:

{
  "mastery": {
    "topic": "machine_learning",
    "mastery_level": 0.785,
    "semantic_avg": 0.802,
    "keyword_avg": 0.768,
    "sessions_attempted": 12,
    "questions_attempted": 96,
    "correct_count": 78,
    "avg_response_time": 45.2,
    "current_difficulty": "medium",
    "mastery_velocity": 0.042,
    "consecutive_good": 3,
    "consecutive_poor": 0,
    "weak_concepts": ["neural_networks", "backpropagation"],
    "strong_concepts": ["regression", "classification"],
    "concept_stagnation": {"gradient_descent": 2}
  },
  "exists": true
}

Located in app.py:2081-2095

Understanding Metrics

The platform tracks several key performance indicators:

Semantic Similarity

What it measures: How well your answers match expected responses in meaning and content. How it’s calculated:

Uses SentenceTransformer embeddings (all-MiniLM-L6-v2)
Computes cosine similarity between your answer and expected answer
Range: 0.0 (no match) to 1.0 (perfect match)

What the scores mean:

0.9-1.0: Excellent understanding, comprehensive answer
0.7-0.89: Good grasp, minor details missing
0.5-0.69: Partial understanding, needs improvement
Below 0.5: Weak understanding, requires focused study

Field: semantic_avg - Average across all questions in topic

Keyword Coverage

What it measures: Presence of important technical terms and concepts in your answers. How it’s calculated:

Extracts key terms from expected answer
Checks which terms appear in your response
Percentage of expected keywords found

Example:

Expected keywords: ["supervised", "training", "labels", "algorithm"]
Your answer contains: ["supervised", "training", "model"]
Keyword coverage: 2/4 = 0.50

Field: keyword_avg - Average keyword coverage score

Mastery Level

What it measures: Your overall proficiency in a topic. How it’s calculated:

Combines semantic similarity and keyword coverage
Weighted by recent performance (recent answers count more)
Adjusted based on question difficulty

Formula (simplified):

mastery_level = (semantic_avg * 0.6) + (keyword_avg * 0.4)

Field: mastery_level - Overall topic mastery (0-1) Interpretation:

0.85+: Advanced mastery, ready for senior-level questions
0.70-0.84: Solid understanding, intermediate level
0.50-0.69: Developing skills, needs practice
Below 0.50: Beginner level, focus on fundamentals

Learning Velocity

What it measures: How quickly you’re improving. How it’s calculated:

mastery_velocity = current_mastery - last_mastery

Field: mastery_velocity Interpretation:

Positive: You’re improving
Zero: Performance plateau
Negative: Recent performance decline (may indicate fatigue or harder questions)

Response Time

What it measures: Average time to answer questions. Field: avg_response_time (in seconds) Usage:

Tracks your thinking speed
Identifies topics where you hesitate
Helps calibrate interview readiness

Session History and Progress

View All Sessions

Retrieve your complete session history:

GET /api/dashboard_data
Authorization: Bearer <your-token>

Includes:

All interview sessions (mock, technical, coding)
Debugging sessions
Scores and completion dates
Session-specific feedback

Filtering:

GET /api/sessions?type=technical

Filter by session type to focus on specific practice areas. Located in app.py:3170-3186

Recent Performance

Get your last 20 sessions sorted by date:

GET /api/sessions
Authorization: Bearer <your-token>

Response:

{
  "sessions": [
    {
      "id": 145,
      "session_type": "technical",
      "topic": "machine_learning",
      "score": 78.5,
      "duration": 600,
      "created_at": "2026-03-03T14:30:00",
      "completed_at": "2026-03-03T14:40:00"
    }
  ]
}

Performance Trends

Track improvement over time: Sessions Attempted:

Shows practice frequency
Located in sessions_attempted field

Correct Count vs. Questions Attempted:

Accuracy rate = correct_count / questions_attempted
Indicates answer quality

Mastery Velocity:

Positive trend = consistent improvement
Negative trend = may need to review fundamentals

Concept-Level Tracking

The platform tracks mastery at the individual concept level within each topic.

Concept Masteries

Detailed data for each concept stored as JSON:

concept_masteries = {
    "neural_networks": {
        "mastery_level": 0.65,
        "attempts": 8,
        "correct": 5,
        "is_weak": True,
        "is_strong": False,
        "stagnation_count": 1,
        "last_seen": 1709478600.0
    },
    "linear_regression": {
        "mastery_level": 0.92,
        "attempts": 12,
        "correct": 11,
        "is_weak": False,
        "is_strong": True,
        "stagnation_count": 0
    }
}

Access methods:

# Get all concept data
mastery.get_concept_masteries()

# Get weak concepts (mastery < 0.6)
mastery.get_weak_concepts()

# Get strong concepts (mastery >= 0.8)
mastery.get_strong_concepts()

# Get stagnation counts
mastery.get_concept_stagnation()

Located in models.py:112-147

Weak vs. Strong Concepts

Weak Concepts (is_weak: true):

Mastery level below 0.6
Require focused practice
Appear more frequently in adaptive sessions

Strong Concepts (is_strong: true):

Mastery level above 0.8
Well understood
Appear less frequently to avoid redundancy

Concept Stagnation

What it tracks: Concepts where mastery isn’t improving despite practice. Field: stagnation_count

Increments when mastery doesn’t increase after attempts
Used to adjust teaching strategy
May trigger different question types or explanations

Adaptive Difficulty

The system automatically adjusts question difficulty based on your performance.

Difficulty Levels

Field: current_difficulty Values:

easy: Foundational questions
medium: Standard interview questions (default)
hard: Advanced and complex scenarios

Difficulty Progression

Triggered by streaks: consecutive_good (good performances):

3+ consecutive good answers → increase difficulty
Keeps you challenged and engaged

consecutive_poor (poor performances):

3+ consecutive weak answers → decrease difficulty
Helps rebuild confidence and fundamentals

Reset conditions:

Streaks reset when you change difficulty levels
Reset when switching topics

Resetting Mastery

If you want to start fresh or practice from the beginning:

Reset Specific Topic

POST /api/user/reset_mastery
Authorization: Bearer <your-token>
{
  "topic": "machine_learning"
}

Effect:

Deletes UserMastery record for that topic
Clears all concept data
Preserves session history

Reset All Topics

POST /api/user/reset_mastery
Authorization: Bearer <your-token>
{
  "topic": "all"
}

Effect:

Deletes all UserMastery records
Fresh start across all topics
Session history remains intact

Located in app.py:2234-2279

Resetting mastery is irreversible. Your session history is preserved, but all progress metrics and concept tracking will be deleted.

Debugging Sessions

For debugging practice, separate tracking is available:

GET /api/profile/debugging_history
Authorization: Bearer <your-token>

Includes:

Debugging session records
Code challenges completed
Performance summaries
Concepts practiced

Located in app.py:2801-2806

Action Plans

Get personalized learning recommendations:

GET /api/profile/action_plans
Authorization: Bearer <your-token>

Based on your mastery data, receives:

Topics to focus on
Specific weak concepts to practice
Recommended session types
Difficulty adjustments

Located in app.py:2915-2932

Best Practices

Maximize your learning:

Practice regularly: Consistent practice improves mastery velocity
Review weak concepts: Focus on areas with low mastery levels
Monitor stagnation: If a concept isn’t improving, try different learning resources
Track velocity: Positive velocity means your study approach is working
Use session history: Review past feedback to avoid repeating mistakes

Understanding your metrics:

Semantic similarity reflects comprehension depth
Keyword coverage shows technical vocabulary knowledge
Combined mastery gives the complete picture
Response time indicates confidence level

Next Steps

Resume Practice

Continue with more interview sessions

Resume Analysis

Find skill gaps to address

Interview Endpoints

Explore progress tracking endpoints

Adaptive Learning

View comprehensive mastery data

Overview

Getting Started

Core Features

User Guide

Overview

UserMastery Model

Viewing Mastery Levels

Overall Progress

Topic-Specific Mastery

Understanding Metrics

Semantic Similarity

Keyword Coverage

Mastery Level

Learning Velocity

Response Time

Session History and Progress

View All Sessions

Recent Performance

Performance Trends

Concept-Level Tracking

Concept Masteries

Weak vs. Strong Concepts

Concept Stagnation

Adaptive Difficulty

Difficulty Levels

Difficulty Progression

Resetting Mastery

Reset Specific Topic

Reset All Topics

Debugging Sessions

Action Plans

Best Practices

Next Steps

Resume Practice

Resume Analysis

Interview Endpoints

Adaptive Learning

Build docs developers (and LLMs) love

Overview

Getting Started

Core Features

User Guide

​Overview

​UserMastery Model

​Viewing Mastery Levels

​Overall Progress

​Topic-Specific Mastery

​Understanding Metrics

​Semantic Similarity

​Keyword Coverage

​Mastery Level

​Learning Velocity

​Response Time

​Session History and Progress

​View All Sessions

​Recent Performance

​Performance Trends

​Concept-Level Tracking

​Concept Masteries

​Weak vs. Strong Concepts

​Concept Stagnation

​Adaptive Difficulty

​Difficulty Levels

​Difficulty Progression

​Resetting Mastery

​Reset Specific Topic

​Reset All Topics

​Debugging Sessions

​Action Plans

​Best Practices

​Next Steps

Resume Practice

Resume Analysis

Interview Endpoints

Adaptive Learning

Build docs developers (and LLMs) love

Overview

UserMastery Model

Viewing Mastery Levels

Overall Progress

Topic-Specific Mastery

Understanding Metrics

Semantic Similarity

Keyword Coverage

Mastery Level

Learning Velocity

Response Time

Session History and Progress

View All Sessions

Recent Performance

Performance Trends

Concept-Level Tracking

Concept Masteries

Weak vs. Strong Concepts

Concept Stagnation

Adaptive Difficulty

Difficulty Levels

Difficulty Progression

Resetting Mastery

Reset Specific Topic

Reset All Topics

Debugging Sessions

Action Plans

Best Practices

Next Steps