Interview Copilot Platform

A RAG-based interview assistant platform that generates accurate answers during interview processes, similar to ntro.io.

Features

1. Conversation History Answers: Generates context-aware answers based on previous conversation
2. Resume-Based Knowledge Base: Uses interviewee's resume and profile to provide personalized answers
3. Pre-Prepared Answer Vault: Provides curated answers for difficult questions (STAR stories, playbooks)
4. Continuous Learning: Improves through recommendation system based on interview data

Architecture

Runtime Architecture

See runtime-diagram.md for the complete runtime flow diagram.

Offline Improvement Loop

See offline-diagram.md for the continuous learning system.

Complete Documentation

See complete-architecture.md for detailed system design and implementation guide.

Quick Start

Prerequisites

• Python 3.10+
• Node.js 18+
• PostgreSQL (or SQLite for development)
• OpenAI API key (or Anthropic API key, or use free sentence-transformers)

Backend Setup

1. Navigate to backend:

cd backend

2. Create virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r ../requirements.txt

4. Configure environment:

# Create .env file in backend/ directory
cp .env.example .env
# Edit .env and add your API keys

5. Run the backend:

python main.py

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

Frontend Setup

1. Navigate to frontend:

cd frontend

2. Install dependencies:

npm install

3. Run the frontend:

npm run dev

The frontend will be available at http://localhost:3000

Project Structure

.
├── backend/
│   ├── app/
│   │   ├── api/          # API endpoints
│   │   ├── core/          # Configuration and database
│   │   ├── models/        # Database models
│   │   └── services/      # Business logic services
│   └── main.py           # FastAPI application
├── frontend/
│   ├── src/
│   │   ├── components/   # React components
│   │   └── App.jsx       # Main app component
│   └── package.json
├── runtime-diagram.md    # Runtime architecture diagram
├── offline-diagram.md    # Offline improvement diagram
├── complete-architecture.md  # Full documentation
└── requirements.txt      # Python dependencies

API Documentation

Once the backend is running:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc

Key Components

Knowledge Sources

1. Session Memory: Conversation history with timestamps
2. Resume & Profile KB: User-specific information
3. Prepared Answer Vault: STAR stories and playbooks
4. Company/Role Pack: Job-specific context

Core Services

• Router: Intelligently routes queries to knowledge sources
• Embedding Service: Converts text to vectors
• Vector Store: Manages ChromaDB collections
• LLM Service: Generates answers using GPT-4/Claude
• Guardrails: Validates and filters outputs
• Recommendation Engine: Analyzes logs for improvements

Usage Example

1. Create a user
2. Upload resume
3. Add prepared answers
4. Start an interview session
5. Ask questions and get AI-generated suggestions
6. System learns from your actions

Technology Stack

• Backend: FastAPI, SQLAlchemy, ChromaDB
• Frontend: React, Vite
• LLM: OpenAI GPT-4 or Anthropic Claude
• Embeddings: OpenAI or sentence-transformers
• Vector DB: ChromaDB (can be swapped for Pinecone, Weaviate, etc.)

License

[Add your license here]

Contributing

[Add contribution guidelines]