MAIA - Mental Health AI Assistant

Your compassionate AI companion for mental wellness

MAIA is a web-based AI-powered mental health support platform that provides 24/7 emotional support in a safe, private space. It features a patient-facing AI chat interface and a therapist-facing dashboard for professional oversight.

⚠️ Important Disclaimer: MAIA does not diagnose conditions, replace professional therapy, or provide medical advice. It is designed as a supportive tool to complement professional mental health care.

✨ Features

AI-Powered Capabilities (via Backboard)

MAIA leverages Backboard to provide advanced AI features (Some Detailed Flows):

🧠 Stateful Memory

• Persistent Conversations - MAIA remembers context across sessions, maintaining continuity in therapeutic conversations
• Per-User Assistants - Each patient gets a dedicated AI assistant with isolated memory and context

📚 RAG (Retrieval-Augmented Generation)

• CBT Knowledge Base - AI responses are grounded in Cognitive Behavioral Therapy literature and best practices
• Therapist Notes Integration - Therapists can upload patient notes that inform MAIA's responses for personalized support

🛠️ Custom Tools

• Guardian AI - Real-time detection of concerning patterns or crisis indicators, automatically notifying therapists

For Patients

• 💬 Empathetic AI Chat - 24/7 access to a supportive AI companion that remembers your journey
• 📜 Chat History - Full access to previous conversations
• 🧘 CBT-Informed Support - Responses guided by evidence-based therapeutic techniques
• 👨‍⚕️ Therapist Connection - View assigned therapist information
• 🔗 Assignment Management - Accept or deny therapist assignment requests

For Therapists

• 📊 Patient Dashboard - Centralized view of all assigned patients
• 📝 Session Notes - Upload and manage patient notes that enhance AI personalization
• 📋 Weekly Summary Generation - AI-generated summaries of patient conversations and progress
• 🚨 Safety Alerts - Real-time notifications when the AI detects concerning patterns or crisis indicators (custom tool integration)
• 📧 Patient Assignment - Request assignment to patients via email

🛠️ Tech Stack

Backend

• Framework: FastAPI (Python 3.13+)
• Database: SQLite with SQLAlchemy (async)
• AI Integration: Backboard SDK
• Authentication: JWT (PyJWT) with OAuth2
• Password Hashing: Argon2 (pwdlib)

Frontend

• Framework: Next.js 15 (App Router)
• UI Components: Radix UI primitives
• Styling: Tailwind CSS
• Forms: React Hook Form with Zod validation
• Icons: Lucide React

🚀 Getting Started

Prerequisites

• Python 3.13 or higher
• Node.js 18 or higher
• npm, yarn, pnpm, or bun

Backend Setup

1. Navigate to the backend directory:

   cd backend

2. Create a virtual environment and install dependencies:

   # Using uv
   uv sync
   

3. Set up environment variables:

   # Create a .env file in the backend directory
   echo "BACKBOARD_API_KEY=your_api_key_here" > .env

4. Run the development server:

   uv run uvicorn api.main:app --reload

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

Frontend Setup

1. Navigate to the frontend directory:

   cd frontend

2. Install dependencies:

   npm install
   # or
   yarn install
   # or
   pnpm install

3. Set up environment variables (optional):

   # Create a .env.local file
   echo "NEXT_PUBLIC_API_URL=http://127.0.0.1:8000" > .env.local

4. Run the development server:

   npm run dev

   Open http://localhost:3000 in your browser.

Future Improvements

• Find a solution to uploading RAG documents for each user individually
• Fix the bug where threads/assistants get created for therapist accounts

📄 License

This project is under the GNU AFFERO GENERAL PUBLIC LICENSE.

🤝 Contributing

Please open an issue.

---

Made with ❤️ for mental wellness