📧 AI-Powered Email Assistant

An intelligent email assistant built with React (Vite), FastAPI, Redis, and Celery, powered by LLaMA-70B.
This project transforms how users interact with their inbox by combining task automation, smart replies, and natural language querying.

---

🚀 Features

• ✅ Extract Todos from Emails
  Automatically scans incoming emails to identify actionable tasks such as deadlines, meetings, or follow-ups.
  This eliminates the need for manual task tracking and helps users stay organized directly from their inbox.

• 📬 Smart Reply Generator
  Determines whether an email requires a reply and, if so, drafts a contextually relevant and professional response.
  This feature saves time by handling routine replies and ensures consistent communication tone.

• ✍️ AI-Powered Email Composer
  Allows users to draft new emails by conversing with the LLM in natural language.
  Instead of struggling to phrase a message, users can describe their intent (e.g., “Write a polite follow-up about yesterday’s meeting”) and receive a polished email instantly.

• 🔍 Inbox Q&A
  Provides a conversational interface to query the inbox using plain English.
  For example:

  • “Show me all emails from my advisor last week.”
  • “What meetings are scheduled tomorrow?”
  • “Summarize unread emails from today.”
    This reduces time spent searching manually and makes email management more intuitive.

---

🛠️ Tech Stack

• Frontend: React + Vite
• Backend: Python + FastAPI + Uvicorn
• AI Model: LLaMA 80B
• Task Queue: Celery + Redis

---

⚙️ Installation & Setup

Prerequisites

• Node.js (v18+)
• Python (3.10+)
• Redis server running locally

1. Clone the Repository

git clone https://github.com/your-username/mailmind.git cd mailmind

2. Frontend Setup

cd frontend npm install npm run dev # starts Vite dev server

3. Backend Setup

cd backend pip install -r requirements.txt uvicorn main:app --reload

4. Start Celery Worker

celery -A worker.celery_app worker --loglevel=info