A full-stack AI-powered mental health support application that provides evidence-based therapeutic guidance through natural language conversations. Built with React, Django, and LlamaIndex for intelligent document retrieval.

• AI-Powered Therapy Conversations: Leverages Groq (Llama 3.3 70B) and LlamaIndex for context-aware, evidence-based responses
• Multiple Therapy Modalities: Supports various therapy types including CBT, DBT, Family Therapy, and Teen Counseling
• Crisis Detection: Automatically detects crisis messages and directs users to appropriate resources
• Knowledge Base Integration: Built on curated mental health resources from trusted sources
• Modern React Frontend: Clean, responsive Material-UI interface
• Django REST API: Robust backend with RESTful endpoints and JWT authentication
• Vector Search: Semantic search through therapy documents using remote HuggingFace embeddings

TherapyBot/
├── backend/
│   ├── djangoProject/        # Django project config
│   │   ├── settings.py       # Django configuration
│   │   ├── urls.py           # URL routing
│   │   └── wsgi.py           # WSGI entrypoint
│   ├── therapybot/           # Django app
│   │   ├── views.py          # API endpoints
│   │   └── urls.py           # App URL routing
│   ├── testingAI.py          # LlamaIndex + Groq integration
│   ├── storage/              # Vector index storage
│   └── manage.py
├── frontend/
│   ├── src/
│   │   ├── components/       # React components
│   │   ├── api/              # API client
│   │   └── App.tsx           # Main app
│   └── package.json
├── requirements.txt
└── Dockerfile

• Python 3.11+
• Node.js & npm (for React frontend)
• Groq API Key (free)
• HuggingFace API Key (free, Read access only)

1. Clone the repository

   git clone https://github.com/RasheedMartin/TherapyBot.git
   cd TherapyBot

2. Create and activate virtual environment

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

3. Install dependencies

   pip install -r requirements.txt

4. Set up environment variables — create a .env file in the backend directory:

   GROQ_API_KEY=your-groq-api-key
   HF_API_KEY=your-huggingface-api-key
   SECRET_KEY=your-django-secret-key
   DEBUG=True
   

5. Run migrations

   cd backend
   python manage.py migrate

6. Build the knowledge base (first time only)

   python testingAI.py
   # Type 'rebuild' to create the vector index

7. Run the Django server

   python manage.py runserver

1. Navigate to frontend directory

   cd frontend

2. Install dependencies

   npm install

3. Start the development server

   npm run dev

4. Open your browser

   http://localhost:5173
   

The backend is deployed on Railway using Docker.

Required environment variables in Railway:

Railway automatically runs migrations and starts gunicorn on deploy via the Dockerfile CMD.

The frontend is deployed on Vercel. Connect your GitHub repo and Vercel handles the rest. Set your backend Railway URL as an environment variable in Vercel for API calls.

1. Select a therapy type from the dropdown (e.g., Family Therapy, CBT)
2. Enter your question or concern in the text area
3. Click "Submit" to get an AI-powered response
4. The system retrieves relevant information from the knowledge base and generates a contextual response

Edit the URLs in build_index_local() in testingAI.py:

documents = loader.load_data(
    urls=[
        "https://childmind.org/article/helping-resistant-teens-into-treatment/",
        "https://childmind.org/article/anxiety-disorders-in-children/",
        # Add your URLs here
    ]
)

Frontend:

• React 18
• Material-UI (MUI)
• TanStack Query (React Query)
• TypeScript
• Axios

Backend:

• Django 5.2
• Django REST Framework
• Django REST Framework SimpleJWT
• LlamaIndex 0.14
• Groq API (Llama 3.3 70B)
• HuggingFace Inference API (remote embeddings)

Infrastructure:

• Railway (backend + PostgreSQL)
• Vercel (frontend)
• Docker

GET /api/get_started/
Parameters:
  - question: string (required)
  - therapy_type: string (optional, default: "general therapy")

Response:
{
  "response": "AI-generated therapeutic guidance..."
}

This application is for informational and educational purposes only. It is NOT a substitute for professional mental health care. If you are experiencing a mental health crisis, please contact:

• 988 Suicide & Crisis Lifeline: Call or text 988
• Crisis Text Line: Text HOME to 741741
• Emergency Services: 911

This project is licensed under the MIT License - see the LICENSE file for details.

Rasheed Martin

• GitHub: @RasheedMartin

• Mental health resources from Child Mind Institute
• Built with LlamaIndex
• Powered by Groq

Note: Originally created during a 9-hour hackathon and evolved into a full-stack production application with AI capabilities.