Getting Started

This guide will help you set up and deploy your first AI model on Staque IO.

Prerequisites

Before you begin, ensure you have:

  • Node.js 18.x or higher installed
  • PostgreSQL database (version 13 or higher)
  • AWS account with appropriate IAM permissions
  • OpenAI API key (for AI recommendations feature)
  • NVIDIA API key (optional, for NVIDIA NIM models)

Environment Setup

Create a .env.local file in the root directory with the following variables:

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

# Authentication
JWT_SECRET=your-secret-key-here

# AWS Credentials
STAQUE_AWS_REGION=eu-north-1
STAQUE_AWS_ACCESS_KEY_ID=your-access-key-id
STAQUE_AWS_SECRET_ACCESS_KEY=your-secret-access-key

# SageMaker Configuration (Optional)
SAGEMAKER_SUBNET_IDS=subnet-xxx,subnet-yyy
SAGEMAKER_SECURITY_GROUP_IDS=sg-xxx
SAGEMAKER_EXECUTION_ROLE_ARN=arn:aws:iam::xxx:role/SageMakerRole

# OpenAI (for AI recommendations)
OPENAI_API_KEY=sk-xxx

# NVIDIA NIM (Optional)
NVIDIA_API_KEY=nvapi-xxx
NIM_BASE_URL=https://integrate.api.nvidia.com

Installation

Follow these steps to install and run the application:

1. Clone the Repository

git clone https://github.com/your-org/staque-io.git
cd staque-io

2. Install Dependencies

npm install

3. Set Up Database

Run the database migrations to create the required tables:

npm run db:migrate

Or manually execute the SQL schema from database/schema.sql

4. Run Development Server

npm run dev

The application will be available at http://localhost:3000

First Steps

1. Create an Account

Navigate to /login and create a new user account. The first user will be assigned admin privileges.

2. Configure AWS Credentials

Go to your dashboard and verify that AWS credentials are properly configured by checking the AWS Credentials section in the admin panel.

3. Browse Available Models

Use the "Get Started" page to:

  • Browse available models from AWS Bedrock
  • View SageMaker model packages
  • Explore NVIDIA NIM hosted models

4. Deploy Your First Model

Select a model and follow the deployment wizard:

  1. Choose your model (e.g., Amazon Nova Pro or Claude 3 Sonnet)
  2. Configure deployment settings
  3. Review pricing estimates
  4. Deploy and start chatting

Quick Start Example

Here's a quick example of deploying an AWS Bedrock model using the API:

// Create a conversation with deployed resource
const response = await fetch('/api/conversations', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer YOUR_JWT_TOKEN'
  },
  body: JSON.stringify({
    title: 'My First AI Assistant',
    use_case: 'general-purpose',
    deployed_resource: {
      resource_name: 'Nova Pro Assistant',
      resource_type: 'bedrock',
      aws_resource_id: 'amazon.nova-pro-v1:0',
      region: 'eu-north-1',
      estimated_hourly_cost: 0
    }
  })
})

const { conversation_id, resource_id } = await response.json()

// Start chatting
const chatResponse = await fetch('/api/chat/thread', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'Authorization': 'Bearer YOUR_JWT_TOKEN'
  },
  body: JSON.stringify({
    message: 'Hello! How can you help me?',
    conversationId: conversation_id,
    resourceId: resource_id
  })
})

const { messages } = await chatResponse.json()
console.log(messages)

AWS IAM Permissions

Your AWS IAM user/role needs the following permissions:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "bedrock:ListFoundationModels",
        "bedrock:GetFoundationModel",
        "bedrock:InvokeModel",
        "bedrock-runtime:InvokeModel"
      ],
      "Resource": "*"
    },
    {
      "Effect": "Allow",
      "Action": [
        "sagemaker:ListModelPackages",
        "sagemaker:DescribeModelPackage",
        "sagemaker:CreateModel",
        "sagemaker:CreateEndpointConfig",
        "sagemaker:CreateEndpoint",
        "sagemaker:DescribeEndpoint",
        "sagemaker:UpdateEndpoint",
        "sagemaker:DeleteEndpoint",
        "sagemaker:InvokeEndpoint"
      ],
      "Resource": "*"
    },
    {
      "Effect": "Allow",
      "Action": [
        "pricing:GetProducts"
      ],
      "Resource": "*"
    }
  ]
}

Next Steps

Explore Architecture

Learn about the system architecture and design patterns

API Reference

Detailed documentation of all API endpoints

Deployment Guide

Advanced deployment strategies for production

Chat System

Understanding the conversation and threading system

⚠️ Production Considerations

Before deploying to production, ensure you:

  • Use strong JWT secrets and rotate them regularly
  • Enable SSL/TLS for all connections
  • Set up proper database backups
  • Configure rate limiting on API endpoints
  • Review and restrict AWS IAM permissions
  • Enable CloudWatch monitoring and logging