finalround_interview_project

AhanaS07/finalround_interview_project

3.2

If you are the rightful owner of finalround_interview_project and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

The Interview Analytics Platform is a comprehensive data analytics solution designed to analyze user behavior, product engagement, and subscription metrics across various interview preparation products.

Tools
1
Resources
0
Prompts
0

Interview Analytics Platform

A comprehensive data analytics platform for analyzing user behavior, product engagement, and subscription metrics across interview preparation products. Built with Python, SQLite, and modern web technologies.

Overview

This platform provides real-time analytics and insights for a suite of interview preparation products including:

  • Mock Interviews
  • Live Interviews
  • Auto-Apply Features

The system processes user data, session logs, and engagement metrics to deliver actionable business intelligence through interactive dashboards, REST APIs, and conversational AI interfaces.

Features

Core Analytics

  • User Engagement Analysis: Track activation, retention, and product usage patterns
  • Subscription Analytics: Monitor conversion rates across four subscription tiers (none, normal, plus, god)
  • Revenue Metrics: Calculate MRR, ARR, and revenue by plan
  • Product Performance: Analyze engagement by product page and feature
  • Cohort Analysis: Track user behavior by signup cohorts
  • Geographic Distribution: Analyze user demographics and spending patterns

Interfaces

  1. Interactive Dashboard (Streamlit)

    • Real-time metrics visualization
    • Product engagement analysis
    • User demographics and spending insights
    • Session analytics

  2. REST API (FastAPI)

    • Query analytics data programmatically
    • WebSocket support for real-time updates
    • JSON-based response format

  3. Conversational AI Interface (MCP Server)

    • Natural language queries for analytics
    • AI-powered insights and recommendations
    • Integration with AI assistants

Architecture

├── Data Layer
│   ├── SQLite database (interview_data.db)
│   ├── Users table (demographics, subscriptions)
│   ├── Sessions table (user activity)
│   └── Product visits table (engagement data)
│
├── Processing Layer
│   ├── data_loader.py - ETL pipeline
│   ├── revenue_calculator.py - Business metrics
│   ├── user_updater.py - Data enrichment
│   └── analytics_module.py - Core analytics engine
│
└── Presentation Layer
    ├── dashboard.py - Streamlit UI
    ├── api.py - REST API
    └── mcp_server.py - AI conversation interface

Installation

Prerequisites

  • Python 3.11 or higher
  • pip package manager

Setup

  1. Clone the repository:
git clone <repository-url>
cd finalround_interview_project
  1. Install dependencies:
# Core analytics and dashboard
pip install -r requirements.txt

# API server (optional)
pip install -r requirements_api.txt
  1. Prepare data files: Place the following files in the data/ directory:
  • users_data.csv - User profile data (~20,000 users)
  • simulated_logs.json - Activity logs (~30,000 records)

Usage

1. Initialize Database and Run ETL Pipeline

Process raw data and populate the database:

python main.py

This will:

  • Create database schema
  • Load user data from CSV
  • Process session logs
  • Generate revenue metrics
  • Create analytics reports

Output: analysis_report.json with detailed metrics

2. Launch Interactive Dashboard

# Using the shell script
./run_dashboard.sh

# Or directly with Streamlit
streamlit run dashboard.py

Access the dashboard at http://localhost:8501

Features:

  • Overview metrics (users, sessions, revenue)
  • Product engagement visualization
  • User demographics analysis
  • Session analytics
  • Data quality monitoring

3. Start REST API Server

python api.py

API available at http://localhost:8000

Endpoints:

  • GET / - API status and documentation
  • POST /query - Query analytics data
  • POST /metrics - Get specific metrics
  • WebSocket /ws - Real-time updates

Example queries:

# Get conversion rate
curl -X POST http://localhost:8000/query \
  -H "Content-Type: application/json" \
  -d '{"query": "What is the conversion rate for live interviews?"}'

# Get user metrics
curl -X POST http://localhost:8000/metrics \
  -H "Content-Type: application/json" \
  -d '{"metric_type": "user_engagement"}'

4. Launch AI Conversation Interface (MCP Server)

./start_mcp.sh

The MCP server provides:

  • Natural language analytics queries
  • AI-powered insights
  • Integration with AI assistants
  • Context-aware recommendations

Data Schema

Users Table

  • user_id - Unique identifier
  • name, email, country - Profile information
  • age, sex, married - Demographics
  • current_plan - Subscription tier (none/normal/plus/god)
  • total_spent_local - Lifetime spending
  • last_seen, joined_at - Activity timestamps

Sessions Table

  • session_id - Unique session identifier
  • user_id - Foreign key to users
  • duration_minutes - Session length
  • timestamp - Session start time

Product Visits Table

  • visit_id - Unique visit identifier
  • session_id - Foreign key to sessions
  • page_type - Product page visited
  • duration_minutes - Time on page
  • timestamp - Visit timestamp

Key Metrics

Subscription Tiers

  • None: Free tier users
  • Normal: Basic paid subscription
  • Plus: Premium subscription
  • God: Ultimate tier with all features

Product Pages

  • mock_interview: Practice interview sessions
  • live_interview: Real-time interview coaching
  • auto_apply: Automated job application service

Calculated Metrics

  • Monthly Recurring Revenue (MRR)
  • Annual Recurring Revenue (ARR)
  • Conversion rates by product
  • User engagement rate
  • Average session duration
  • Product usage frequency
  • Cohort retention rates

Testing

API Tests

python test_api.py

Chat Client Tests

python test_chat.py

MCP Server Tests

python test_mcp.py

Dashboard Verification

python verify_dashboard.py

Configuration

Environment Variables

Create a .env file:

DATABASE_URL=sqlite:///interview_data.db
API_PORT=8000
DASHBOARD_PORT=8501

MCP Configuration

Edit mcp_config.json to customize the AI conversation interface:

{
  "server_name": "interview-analytics",
  "version": "1.0.0",
  "capabilities": ["query", "analytics", "insights"]
}

Project Structure

finalround_interview_project/
├── data/                          # Data files
│   ├── users_data.csv
│   └── simulated_logs.json
├── main.py                        # ETL pipeline orchestrator
├── database_schema.py             # Database schema and management
├── data_loader.py                 # Data loading utilities
├── data_validation.py             # Data quality checks
├── user_updater.py                # User data enrichment
├── revenue_calculator.py          # Revenue and business metrics
├── analytics_module.py            # Core analytics engine
├── analytics_summary.py           # Summary statistics
├── dashboard.py                   # Streamlit dashboard
├── api.py                         # FastAPI REST API
├── mcp_server.py                  # AI conversation interface
├── chat_client.py                 # Chat client implementation
├── requirements.txt               # Python dependencies
├── requirements_api.txt           # API-specific dependencies
└── run_dashboard.sh              # Dashboard launcher script

Performance

  • Data Processing: ~2-5 seconds for 20,000+ users
  • Query Performance: <100ms for most analytics queries
  • Dashboard Load Time: <2 seconds initial load
  • API Response Time: <50ms average

Troubleshooting

Database Issues

# Remove existing database and rebuild
rm interview_data.db
python main.py

Port Conflicts

If default ports are in use, specify alternatives:

# Dashboard on custom port
streamlit run dashboard.py --server.port 8502

# API on custom port
uvicorn api:app --port 8001

Missing Dependencies

# Reinstall all dependencies
pip install -r requirements.txt --force-reinstall

License

MIT License - feel free to use this code for learning and development purposes.