Auto-QA and Coaching Insights

An AI-powered system that evaluates 100% of customer support calls, scores them across quality dimensions, and generates agent-level + city-level coaching insights with supervisor flagging for high-risk calls.

🏗️ Architecture Overview

┌─────────────────┐     WebRTC      ┌──────────────────┐
│  React Frontend │◄───────────────►│  LiveKit Server  │
│  (Customer UI)  │                 │  (Docker)        │
└────────┬────────┘                 └────────┬─────────┘
         │                                   │
         │ REST API                          │ Recording
         ▼                                   ▼
┌─────────────────┐                 ┌──────────────────┐
│  FastAPI Backend│                 │  S3 / Local      │
│  (Room Tokens)  │                 │  Storage         │
└─────────────────┘                 └──────────────────┘

🚀 Quick Start (Local Development)

Prerequisites

• Docker & Docker Compose - For running LiveKit server
• Python 3.11+ - For backend API
• Node.js 18+ - For frontend

1. Start LiveKit Server

docker-compose up -d

This starts the LiveKit WebRTC server on localhost:7880.

2. Start Backend API

cd backend

# Create virtual environment
python -m venv venv
venv\Scripts\activate  # Windows
# source venv/bin/activate  # Linux/Mac

# Install dependencies
pip install -r requirements.txt

# Run the server
uvicorn app.main:app --reload --host 0.0.0.0 --port 8000

API will be available at http://localhost:8000

3. Start Frontend

cd frontend

# Install dependencies
npm install

# Run dev server
npm run dev

Frontend will be available at http://localhost:3000

📁 Project Structure

Auto-QA-and-Coaching-Insights/
├── backend/                    # FastAPI Backend
│   ├── app/
│   │   ├── main.py            # Application entry point
│   │   ├── config.py          # Environment configuration
│   │   ├── schemas.py         # Pydantic models
│   │   ├── routes/
│   │   │   └── rooms.py       # Room management endpoints
│   │   └── services/
│   │       └── livekit_service.py  # LiveKit integration
│   ├── requirements.txt
│   └── .env
│
├── frontend/                   # React Frontend
│   ├── src/
│   │   ├── App.jsx
│   │   ├── main.jsx
│   │   ├── api/
│   │   │   └── roomService.js # API client
│   │   └── components/
│   │       ├── LandingPage.jsx # Customer landing page
│   │       └── CallRoom.jsx    # WebRTC call interface
│   ├── package.json
│   └── vite.config.js
│
├── docs/                       # Documentation
│   ├── problem_statement.txt
│   └── Technical_srs.txt
│
└── docker-compose.yml          # LiveKit server config

🔌 API Endpoints

Create Room Request

{
  "customer_name": "John Doe",
  "customer_phone": "+1234567890"
}

Create Room Response

{
  "room_name": "support-20260131-120000-abc12345",
  "token": "eyJhbGciOiJIUzI1NiIs...",
  "livekit_url": "ws://localhost:7880",
  "created_at": "2026-01-31T12:00:00Z"
}

🛠️ Development Status

Phase 1: Core Infrastructure ✅

• Project structure setup
• LiveKit Docker configuration
• Backend API for room management
• React landing page with "Help" button
• WebRTC call interface
• AI Voice Agent (Pipecat) - Next

Phase 2: QA Pipeline 🔜

• Call recording (LiveKit Egress)
• AWS Transcribe integration
• Bedrock LLM analysis
• DynamoDB storage
• SNS alerts

Phase 3: Dashboard 🔜

• QA scorecards
• Agent performance trends
• City-level insights
• Supervisor alerts UI

📄 License

Internal use only - Battery Smart Hackathon 2026