Skip to content

Latest commit

 

History

History
127 lines (100 loc) · 5.62 KB

README.md

File metadata and controls

127 lines (100 loc) · 5.62 KB

AWS Coder Workshop Templates

This directory contains Coder workspace templates designed for AWS development workflows and workshop scenarios. Each template provides a specialized development environment with pre-configured tools and services.

Template Overview

☸️ Kubernetes with Kiro CLI

Purpose: AI-powered development with Kiro CLI and MCP server support
Architecture: Kubernetes pods with persistent volumes
Key Tools: Kiro CLI, AWS CLI v2, AWS CDK, Node.js 20 LTS, uv/uvx for MCP servers
Best For: AI-assisted development, infrastructure as code, MCP server integration

🚀 EC2 (Linux) SAM

Purpose: Serverless application development
Architecture: Ubuntu ARM64 on Graviton EC2 instances
Key Tools: AWS SAM CLI, AWS CLI v2, Python 3, Kiro extension
Best For: Lambda functions, serverless APIs, cost-effective ARM64 development

🪟 EC2 (Windows) DCV

Purpose: Windows development with remote desktop
Architecture: Windows Server 2022 on x86_64 EC2 instances
Key Tools: NICE DCV, VS Code, PowerShell, Windows development stack
Best For: Windows-specific development, GUI applications, .NET development

☸️ Kubernetes with Claude Code

Purpose: Container development with AI task automation
Architecture: Kubernetes pods with persistent volumes
Key Tools: Claude Code 4.7.1, AWS Bedrock, AWS CLI, AWS CDK, Node.js
Best For: AI-driven development, task automation, microservices, container orchestration

🤖 Kubernetes with RAG and Claude Code

Purpose: GenAI RAG application prototyping with vector database
Architecture: Kubernetes pods with Aurora PostgreSQL Serverless v2
Key Tools: Claude Code, pgvector, AWS Bedrock, Streamlit, Python tooling
Best For: AI/ML applications, vector search, RAG prototyping, data science

Template Comparison

Feature K8s Kiro CLI Linux SAM Windows DCV K8s Claude Code K8S RAG Claude Code
Platform Kubernetes Ubuntu ARM64 Windows Server Kubernetes Kubernetes
AI Assistant Kiro CLI + MCP Kiro Extension - Claude Code Claude Code
Primary Use General AWS Dev Serverless Windows Dev Container Dev GenAI/RAG Dev
Cost Efficiency Variable High (ARM64) Higher Variable Variable
Persistence Home directory Full VM Full VM Home directory Home directory
Startup Time ~30-60 sec ~2-3 min ~5-10 min ~30-60 sec ~5-10 min

Getting Started

Prerequisites

  • Coder deployment with AWS provider configured
  • AWS account with appropriate IAM permissions
  • For Kubernetes templates: Existing Kubernetes cluster

Template Selection Guide

Choose Kubernetes with Kiro CLI if you want:

  • AI-powered development assistance with MCP server support
  • Infrastructure as Code with CDK
  • General-purpose AWS development on Kubernetes
  • Fast workspace startup times
  • Configurable MCP servers

Choose Linux SAM if you want:

  • Serverless application development
  • Cost-effective ARM64 instances
  • Lambda function development
  • Local serverless testing

Choose Windows DCV if you want:

  • Windows-specific development
  • GUI application development
  • .NET or Windows-only tools
  • Remote desktop experience

Choose Kubernetes Claude Code if you want:

  • Container-based development
  • AI task automation with Claude Code
  • Microservices architecture
  • Fast workspace startup times

Choose Kubernetes RAG Claude Code if you want:

  • GenAI/RAG application development
  • Vector database integration
  • AI/ML prototyping with Bedrock
  • Streamlit-based data applications

Configuration

IAM Instance Profile

EC2-based templates (Linux SAM, Windows DCV) require an IAM instance profile for AWS service access. Configure the aws_iam_profile variable when deploying templates.

Kubernetes Service Account

Kubernetes-based templates use ServiceAccount with IAM role binding for AWS service access.

Regional Deployment

All templates support multi-region deployment with region-specific optimizations:

  • Kubernetes Kiro CLI: Depends on cluster location
  • Linux SAM: 4 US regions (ARM64 availability)
  • Windows DCV: 15 regions globally
  • Kubernetes Claude Code: Depends on cluster location
  • RAG Claude Code: Depends on cluster location + Aurora availability

Resource Sizing

Each template offers configurable resource options:

  • CPU: 2-8 vCPUs depending on template
  • Memory: 4-16 GiB RAM options
  • Storage: 10-50 GB persistent volumes (Kubernetes) or 10-300 GB (EC2)
  • Database: Aurora Serverless v2 (0.5-1.0 ACU for RAG template)

Workshop Integration

These templates are designed for the AWS Modernization with Coder workshop and include:

  • Pre-configured development environments
  • Workshop-specific tooling and dependencies
  • Optimized resource configurations
  • Integration with workshop exercises

Support and Customization

Each template is designed as a starting point and can be customized for specific use cases:

  • Modify Terraform configurations for additional AWS services
  • Extend startup scripts for custom tool installation
  • Adjust resource parameters for performance requirements
  • Add custom applications and integrations

Note: These templates are designed for workshop and development use. For production deployments, review and adjust security configurations, resource limits, and access controls according to your organization's requirements.