---
title: UW Aya-Expanse-32b
emoji: 💬
colorFrom: yellow
colorTo: purple
sdk: gradio
sdk_version: 5.42.0
app_file: app.py
pinned: false
hf_oauth: true
hf_oauth_scopes:
  - inference-api
license: mit
---

# 🎤 Interview Simulator

An AI-powered multilingual interview simulator built with HuggingFace's Aya-Expanse-32b model.

## Features

- **Multilingual Support**: Conduct interviews in English, Spanish (México), French, and Haitian Creole
- **Customizable Configuration**: Set company name, position, questions, and ending statements
- **Auto-Respond Feature**: Simulate candidate responses for testing
- **EEOC Compliance**: Built-in guidelines for fair hiring practices
- **Dual UI Options**: Available in both Gradio and Streamlit interfaces

## Running Locally

### Prerequisites

```bash
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt
```

### Set HuggingFace Token

```bash
export HF_TOKEN=your_huggingface_token_here
```

### Run Gradio Version

```bash
python app.py
```

Then open http://127.0.0.1:7860 in your browser.

### Run Streamlit Version

```bash
streamlit run streamlit_app.py
```

The Streamlit app will automatically open in your browser.

## Configuration

Both versions support the same configuration options:

- **Company Name**: Your organization name
- **Position/Screener Name**: The job title being interviewed for
- **Questions**: Custom interview questions (one per line)
- **Ending Statement**: Closing message for candidates
- **Model Parameters**: Max tokens, temperature, and top-p sampling

## Files

- `app.py` - Gradio implementation
- `streamlit_app.py` - Streamlit implementation
- `requirements.txt` - Python dependencies

## Notes

Written prototype for speed, not cleanliness. Ignore the immense amount of commits.