Skip to content

An AI Email Intelligence Platform Real-time email intelligence with multi-provider AI fallback, semantic search, OAuth integration. Handles incremental sync and streaming with 70% cold start reduction.

License

Notifications You must be signed in to change notification settings

parbhatkapila4/Vector-Mail

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

VectorMail Logo

VectorMail

The AI Email Client That Actually Saves You Time

Stop drowning in emails. Start being productive.

Live Demo GitHub Stars License TypeScript


Try VectorMail Free Β Β |Β Β  Watch Demo Β Β |Β Β  Documentation


🎯 The Problem We Solve

The average professional spends 28% of their workweek on email. That's 11+ hours searching, reading, writing, and organizing. Time that should go to actual work.

Traditional email clients were built for the 1990s. VectorMail is built for how we work today.


Why VectorMail?

Semantic Search That Actually Works

Search by meaning, not just keywords. Ask "emails about the budget meeting last month" and actually find them. Powered by vector embeddings and pgvector.

AI That Understands Context

Every email gets an intelligent summary, automatic categorization, and smart tagging. Know what's important at a glance without reading everything.

Write Emails in Seconds

Describe what you want to say, and our AI composes it with the right tone, context from previous conversations, and your writing style.

Chat With Your Inbox

"Show me all receipts from last quarter" or "Find the email where John mentioned the deadline." Natural language meets your inbox.


Key Features

🧠 AI-Powered Intelligence
Feature Description
Smart Summaries Every email automatically summarized with key points, action items, and context
Intelligent Tagging AI categorizes emails as urgent, informational, promotional, or action-required
Vector Embeddings 768-dimensional embeddings for each email enable true semantic understanding
Priority Detection Automatically surfaces what matters and deprioritizes noise
πŸ” Next-Gen Search
Feature Description
Semantic Search Find emails by meaning, not exact words
Natural Language Search like you'd ask a colleague
Relevance Scoring Results ranked by actual importance, not just recency
Instant Results Sub-100ms search across thousands of emails
AI Composition
Feature Description
Context-Aware Writing AI reads the thread and writes appropriate responses
Tone Adjustment Professional, casual, or customβ€”match any situation
One-Click Replies Generate complete, thoughtful responses instantly
Smart Suggestions Real-time writing assistance as you type
Productivity Dashboard
Feature Description
Email Analytics Track response times, volume patterns, and productivity
Communication Insights Understand who you email most and when
Action Item Tracking Never miss a follow-up or commitment
Enterprise-Ready Security
Feature Description
Clerk Authentication Enterprise-grade auth with MFA, SSO support
Data Encryption End-to-end encryption for all stored data
Privacy First Your data stays yoursβ€”we don't train on your emails
SOC 2 Ready Built with compliance requirements in mind

Architecture

VectorMail is built on a modern, scalable architecture designed for performance and reliability.

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                              CLIENT LAYER                                    β”‚
β”‚  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”          β”‚
β”‚  β”‚   Next.js   β”‚  β”‚   React 19  β”‚  β”‚  Tailwind   β”‚  β”‚   Framer    β”‚          β”‚
β”‚  β”‚     15      β”‚  β”‚             β”‚  β”‚     CSS     β”‚  β”‚   Motion    β”‚          β”‚
β”‚  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜          β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                                      β”‚
                                      β–Ό
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                               API LAYER                                      β”‚
β”‚  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”     β”‚
β”‚  β”‚                    tRPC (End-to-End Type Safety)                    β”‚     β”‚
β”‚  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜     β”‚
β”‚  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”          β”‚
β”‚  β”‚   Account   β”‚  β”‚   Thread    β”‚  β”‚   Search    β”‚  β”‚     AI      β”‚          β”‚
β”‚  β”‚   Router    β”‚  β”‚   Router    β”‚  β”‚   Router    β”‚  β”‚   Router    β”‚          β”‚
β”‚  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜          β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                                      β”‚
                                      β–Ό
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                            INTELLIGENCE LAYER                                β”‚
β”‚  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”          β”‚
β”‚  β”‚   Gemini    β”‚  β”‚   OpenAI    β”‚  β”‚  Embedding  β”‚  β”‚   Vector    β”‚          β”‚
β”‚  β”‚   2.5 Flash β”‚  β”‚   GPT-4     β”‚  β”‚   Service   β”‚  β”‚   Search    β”‚          β”‚
β”‚  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜          β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                                      β”‚
                                      β–Ό
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                              DATA LAYER                                      β”‚
β”‚  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”          β”‚
β”‚  β”‚ PostgreSQL  β”‚  β”‚   pgvector  β”‚  β”‚   Prisma    β”‚  β”‚   Aurinko   β”‚          β”‚
β”‚  β”‚     16+     β”‚  β”‚  Extension  β”‚  β”‚     ORM     β”‚  β”‚  Email API  β”‚          β”‚
β”‚  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜          β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

πŸ› οΈ Tech Stack

Category Technologies
Frontend Next.js 15, React 19, TypeScript, Tailwind CSS, Framer Motion, Radix UI
Backend tRPC, Prisma ORM, PostgreSQL 16+, pgvector
AI/ML Google Gemini 2.5 Flash, OpenAI GPT-4, Custom Embeddings (768-dim)
Auth Clerk (OAuth, MFA, Session Management)
Email Aurinko API (Google, Microsoft 365)
Testing Jest, React Testing Library, Playwright
DevOps Docker, GitHub Actions

Quick Start

Prerequisites

  • Node.js 20+
  • PostgreSQL 16+ with pgvector extension
  • Package Manager npm, yarn, or bun

1. Clone & Install

git clone https://github.com/parbhatkapila4/Vector-Mail.git
cd Vector-Mail
npm install

2. Configure Environment

cp .env.example .env.local

Add your credentials:

# Database
DATABASE_URL="postgresql://user:password@localhost:5432/vectormail"

# Authentication (Clerk)
NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY="pk_..."
CLERK_SECRET_KEY="sk_..."

# Email Service (Aurinko)
AURINKO_CLIENT_ID="..."
AURINKO_CLIENT_SECRET="..."

# AI Services
OPENROUTER_API_KEY="..."
GEMINI_API_KEY="..."

3. Initialize Database

npm run db:push
npm run db:generate

4. Start Development

npm run dev

Open http://localhost:3000 - your AI email client is ready!


Docker Deployment

docker-compose up -d

This spins up:

  • PostgreSQL with pgvector
  • VectorMail application
  • Auto-configured networking

Available Scripts

Command Description
npm run dev Start development server with Turbopack
npm run build Build for production
npm run start Start production server
npm run lint Run ESLint
npm run typecheck Run TypeScript type checking
npm run test Run unit tests (watch mode)
npm run test:ci Run tests with coverage
npm run test:e2e Run Playwright E2E tests
npm run db:studio Open Prisma Studio

Pricing

Plan Price Features
Basic Free 5 AI summaries/day, basic search, single account
Pro $9.99/mo Unlimited AI, advanced search, 5 accounts, priority support
Enterprise $60/mo Everything + custom AI training, SSO, dedicated support

View Full Pricing β†’


πŸ—ΊοΈ Roadmap

  • Semantic Search - Vector-based email search
  • AI Summaries - Automatic email summarization
  • AI Compose - Context-aware email writing
  • Multi-Account - Support for multiple email accounts
  • Mobile App - iOS & Android native apps
  • Calendar Integration - Smart scheduling from emails
  • Team Workspaces - Shared inboxes & collaboration
  • Plugins/Extensions - CRM, Slack, Notion integrations
  • On-Premise - Self-hosted enterprise deployment

🀝 Contributing

We welcome contributions! VectorMail is open source and community-driven.

# 1. Fork the repo
git clone https://github.com/YOUR_USERNAME/Vector-Mail.git

# 2. Create a feature branch
git checkout -b feature/amazing-feature

# 3. Make your changes and commit
git commit -m "feat: add amazing feature"

# 4. Push and open a PR
git push origin feature/amazing-feature

Please read our Contributing Guide for details on our code of conduct and development process.


License

VectorMail is open-source software licensed under the MIT License.


Support & Community


Email Support

parbhat@parbhat.dev


πŸ› Bug Reports

GitHub Issues


πŸ’‘ Feature Requests

GitHub Discussions


Acknowledgments

Built with incredible open-source technologies:




VectorMail - Email, Reimagined with AI


Built by Parbhat Kapila


Website Β· GitHub Β· Twitter


⭐ Star us on GitHub - it motivates us to keep building!


About

An AI Email Intelligence Platform Real-time email intelligence with multi-provider AI fallback, semantic search, OAuth integration. Handles incremental sync and streaming with 70% cold start reduction.

Topics

Resources

License

Contributing

Stars

Watchers

Forks

Languages