README.md

---
title: RAG Interview Agent
emoji: 🤖
colorFrom: blue
colorTo: purple
sdk: streamlit
app_file: app.py
pinned: true
---

# RAG-Powered AI Interview Agent

An intelligent interview assistant that evaluates candidates using pure Retrieval-Augmented Generation (RAG) architecture.

![Demo](https://i.imgur.com/JQ9w5bG.gif)

## 🚀 Features

- **CV Screening**: Automatic qualification check using semantic similarity
- **Smart Interviews**: Context-aware questions generated from job requirements
- **Fair Evaluation**: Answers scored against knowledge base
- **Detailed Reports**: PDF transcripts with scores and feedback

## 🛠️ Tech Stack

| Component          | Technology Used |
|--------------------|----------------|
| LLM                | Meta Llama-3-8B |
| Vector Store       | FAISS           |
| Embeddings         | Sentence-Transformers |
| UI Framework       | Streamlit       |
| CV Parsing         | PyPDF2, python-docx |

## 📦 Installation

```bash
# Clone the repository
git clone https://huggingface.co/spaces/Jekyll2000/interview_agent
cd your-space-name

# Install dependencies
pip install -r requirements.txt

# Set Hugging Face token
export HUGGINGFACEHUB_API_TOKEN='your-api-token'

# Launch the app
streamlit run app.py

interview-agent/
├── app.py                 # Main application
├── utils/
│   ├── cv_processor.py    # CV parsing and evaluation
│   ├── rag_agent.py       # Core RAG implementation
│   ├── evaluator.py       # Scoring logic
│   └── report_generator.py# PDF report creation
├── data/                  # Job requirements and interviews
└── requirements.txt       # Python dependencies