Agentic AI Solutions

Building intelligent solutions for real-world problems

Documentation

Learn more in the official documentation.

Work in progress!

Installation

⚠️ Important
Make sure the following tools are installed on your computer

🛠️ Quick Start

conda activate curve-ai

1. Initialize the project

pnpm run init

2. Development mode

pnpm run dev

Visit localhost:8099 to view your application

3. Production build

pnpm run build
pnpm run start

📈 Success Metrics

• Functionality and accuracy in task automation
• User adoption and engagement rates
• Integration success with existing workflows
• Client satisfaction and feedback scores

🤝 Join Our Pilot Program

We're currently seeking pilot partners to help validate and refine our solutions. Benefits include:

• Early access to cutting-edge AI solutions
• Customized implementation support
• Reduced pricing for early adopters
• Direct input into product development

📚 Documentation

For detailed information about our solutions and implementation guides, visit our official documentation.

💬 Support

Need help? We're quick to respond:

• Join our Discord community
• Follow us on Twitter @marblismAI

🔒 License

Proprietary software. All rights reserved.

Curve AI Solutions Web App - Deployment Guide

Project Overview

This is a modern web application built with:

• React 18 frontend
• Remix framework for server-side rendering
• Prisma for database ORM
• PostgreSQL database
• TailwindCSS for styling
• Docker for containerization
• ZenStack for access control and API generation
• tRPC for type-safe API calls

Prerequisites for Deployment

• Node.js 20+
• Docker and Docker Compose
• PostgreSQL database
• PNPM package manager

Local Development Setup

Environment Setup (Recommended)

For the best development experience, we recommend using Git Bash with a Conda environment:

1. Create a new conda environment:

   conda create -n curve-ai nodejs=20 -y

2. Activate the environment:

   conda activate curve-ai

3. Install pnpm globally:

   npm install -g pnpm

This approach provides an isolated environment with the correct Node.js version and avoids cross-platform issues between Windows and WSL.

Project Setup

1. Clone the repository:

   git clone https://github.com/claybowl/Curve-Ai-Solutions-Web-App.git
   cd Curve-Ai-Solutions-Web-App

2. Install dependencies:

   pnpm install

3. Create a .env file based on .env.template:

   cp .env.template .env

   Then edit the .env file with your configuration values.

4. Start the development services:

   pnpm run docker:init

5. Generate ZenStack code and initialize the database:

   pnpm run crud:sync
   pnpm run database:sync:dev
   pnpm run database:seed

6. Start the development server:

   pnpm run dev

Troubleshooting Common Issues

ZenStack Schema Generation Issues

If you encounter errors like "Cannot read properties of undefined (reading 'createMany')", run these commands in order:

pnpm run crud:sync
pnpm run database:sync:dev

This regenerates the ZenStack schema and synchronizes the database.

Platform Compatibility Issues

When switching between Windows and WSL:

1. Delete node_modules:

   rm -rf node_modules

2. Reinstall in your current environment:

   pnpm install

This ensures native modules like esbuild are compiled for the correct platform.

Production Deployment Options

1. Traditional Hosting (VPS/Dedicated Server)

1. Set up a VPS with Ubuntu/Debian
2. Install Node.js 20+, Docker, and PostgreSQL
3. Clone the repository
4. Build the application:

   pnpm run build

5. Start with PM2 or similar process manager:

   pm2 start npm --name "curve-ai-app" -- start

2. Platform as a Service (PaaS)

Deploying to Vercel

1. Connect your GitHub repository to Vercel
2. Set the following:
   • Framework Preset: Remix
   • Build Command: pnpm run build
   • Output Directory: build/
   • Install Command: pnpm install
3. Set up environment variables from your .env file
4. Deploy

Deploying to Render

1. Create a new Web Service on Render
2. Connect to your GitHub repository
3. Configure:
   • Build Command: pnpm install && pnpm run build
   • Start Command: pnpm run start
4. Add your environment variables
5. Deploy

3. Docker-based Deployment

This project includes Docker configuration for easy containerized deployment.

1. Build the Docker image:

   docker build -t curve-ai-app .

2. Run the container:

   docker run -p 8099:8099 --env-file .env curve-ai-app

4. Google Cloud Deployment

This application can be deployed on Google Cloud Platform:

1. Database: Use Cloud SQL for PostgreSQL

   • Create a PostgreSQL instance in Google Cloud Console
   • Set up a database named "api"
   • Configure secure access with Cloud SQL Auth Proxy

2. Web App: Deploy using Cloud Run (recommended) or App Engine

   • Cloud Run provides a serverless container environment
   • App Engine offers a managed platform for Node.js apps

3. Environment Variables: Use Secret Manager for sensitive information

4. Google Auth: The existing Google Authentication integration works seamlessly with GCP

Database Considerations

• For production, use a managed PostgreSQL service (AWS RDS, DigitalOcean Managed Databases, etc.)
• Update your DATABASE_URL in the environment variables
• Run migrations before deployment:

  pnpm run database:sync

Environment Variables

Ensure these environment variables are set for production:

• NODE_ENV=production
• BASE_URL - Your production URL
• SERVER_AUTHENTICATION_SECRET - A strong secret key
• SERVER_DATABASE_URL - PostgreSQL connection string
• SERVER_OPENAI_API_KEY - For AI functionality

Monitoring and Maintenance

• Set up application monitoring with services like Sentry or New Relic
• Configure regular database backups
• Implement a CI/CD pipeline for automated deployments

Support

For any deployment issues, please refer to the official documentation or open an issue on GitHub.