GrowEasy CRM - AI Powered Lead Importer

GrowEasy CRM includes an advanced, AI-powered CSV lead importer that takes messy spreadsheets from any source (Facebook Ads, Google Ads, manual exports) and intelligently maps them to structured CRM fields using Large Language Models (LLMs).

Features & Bonus Points Completed

• Intelligent Field Mapping: Extracts fields even if column headers are weird or messy.
• Handling Messy Datasets: Sanitizes data, fixes formats, extracts phone numbers correctly.
• Streaming/Incremental Parsing: Uses csv-parse readable streams for chunked incremental processing. No Out-Of-Memory errors on huge files!
• Retry Mechanism: Implements exponential backoff for failed AI batches to ensure reliable completion.
• Virtualized Tables: Both the Preview and Results tables use @tanstack/react-virtual to smoothly render 100,000+ rows without freezing the browser.
• Real-time Progress Indicators: Visual progress bars and batch-level indicators track AI processing live.
• Drag & Drop Upload: Modern react-dropzone integration for seamless file dropping.
• Dark Mode: Next-themes integration with CSS variables for a stunning dark mode experience.
• Docker Support: Containerized via Docker and docker-compose.
• Unit Tests: Pre-configured with Jest and React Testing Library for frontend and backend logic.
• Modern Tech Stack: Next.js 14 App Router, Express.js backend, and multiple AI providers supported (Gemini, OpenAI, Anthropic).

Architecture

This project is structured as a monorepo using npm workspaces:

• frontend/: Next.js 14 application providing the user interface.
• backend/: Express.js backend server handling file uploads, stream processing, and AI integrations.
• shared/: Shared Typescript types for seamless type safety across the stack.

Setup Instructions

1. Prerequisites

• Node.js v20+
• npm v10+
• A Google Gemini API Key (or OpenAI/Anthropic)

2. Environment Variables

Create a .env file in the backend/ directory:

PORT=3001
GEMINI_API_KEY=your_gemini_key_here
AI_PROVIDER=gemini
AI_MODEL=gemini-1.5-flash

Create a .env.local file in the frontend/ directory (optional):

NEXT_PUBLIC_API_URL=http://localhost:3001

3. Installation & Local Development

Install all dependencies from the root directory:

npm install

Start both the frontend and backend simultaneously:

npm run dev

• Frontend: http://localhost:3000
• Backend: http://localhost:3001

4. Running with Docker

You can spin up the entire stack using Docker Compose:

docker-compose up --build

5. Running Tests

To run the Jest test suites across the monorepo:

npm test

Deployment

Deploying the Frontend (Vercel)

1. Push your code to GitHub.
2. Import the project in Vercel.
3. Set the Root Directory to frontend.
4. Set the Build Command to npm run build.
5. Add the Environment Variable: NEXT_PUBLIC_API_URL pointing to your deployed backend.
6. Deploy!

Deploying the Backend (Render / Railway)

1. Connect your repository to Render/Railway.
2. Set the Root Directory to backend (or run from root with build command npm run build -w shared && npm run build -w backend).
3. Set the Start Command to node dist/index.js (or npm start -w backend).
4. Add all environment variables from your .env file.
5. Deploy!