Implement

Available in product edition: Professional

Workbench supports two implementation approaches for transforming design files into executable code. Both methods use design files as the authoritative source for service architecture and component specifications.

Overview

Design files define the building blocks, relationships and requirements of your service. The implementation approach determines how these specifications are translated into working code:

• Managed Stacks: Automated code generation using predefined technology templates
• Custom Stacks: AI-assisted implementation based on your preferred technology, guided by design specifications

Managed stacks

Managed stacks transform design files into code using deterministic, predefined technology templates. Each stack implements a specific technology combination (e.g., TypeScript/Node.js, Java/Spring Boot) with standardized patterns and configurations.

Characteristics:

• Deterministic transformation: Design-to-code mapping follows predefined rules
• Technology-specific: Each stack supports one technology combination
• Standardized patterns: Consistent implementation patterns across projects
• Automated generation: Minimal manual intervention required

Learn more about managed stacks →

Custom Stacks

Available in product edition: Professional

Custom Stacks enable flexible implementation of services using AI-powered coding assistants. By leveraging design files created with the Solution Builder, this approach provides sustainable and intelligent code generation that adapts to your specific technology choices and architectural patterns.

❗️Beta Feature

This feature is currently in beta and may not be fully optimized yet. We're continuously improving the integration with AI coding assistants and welcome your feedback.

The designed items, reusable components, patterns and architectural decisions construct a coherent design that serves as the foundation for stable code generation. This structured design approach ensures that AI coding assistants can consistently produce reliable, well-architected code that follows your established patterns and standards.

How It Works

Custom Stacks transform your design specifications into working code through AI-powered assistants. Unlike predefined templates, this approach offers complete flexibility in technology selection, implementation patterns and architectural decisions.

Key Advantages:

• Complete Flexibility: Use any programming language supported by your coding assistant
• Architectural Freedom: Implement custom patterns and architectural decisions
• Design-driven Implementation: Code generation follows your design specifications and standards
• Sustainable AI Usage: Design files provide consistent context for reliable AI assistance

Design Files as AI Context

The design files created with the Solution Builder contain all architectural information and serve as comprehensive context for AI coding assistants.

Flexible Implementation Process

1. Design Analysis: AI assistants analyze your design files for components, relationships and patterns
2. Technology Selection: Choose any programming language and framework supported by your coding assistant
3. Pattern Implementation: Implement custom architectural patterns and design decisions
4. Iterative Development: Refine and extend code through ongoing AI assistance
5. Design Compliance: Ensure all generated code maintains design integrity and standards

Supported Tools and Prompts

AI Coding Assistant Integration

Cloud Solution Workbench's design files work seamlessly with AI coding assistants like Cursor IDE, GitHub Copilot and IBM Project Bob using our prompt collection. These prompts are specifically designed to leverage your design files for optimal code generation.

Best Practices

Design File Maintenance

• Keep design files up-to-date with architectural changes
• Ensure design files accurately reflect current system requirements
• Use consistent naming conventions and patterns in design specifications

AI Assistant Usage

• Provide design files as context for all coding sessions
• Use design-specific prompts for consistent results
• Regularly validate generated code against design specifications
• Iterate on implementations while maintaining design compliance

This approach enables you to leverage AI coding assistants in a sustainable, design-compliant manner while maintaining complete flexibility in your implementation choices.

⚠Important Disclaimer

The quality of generated code depends entirely on the AI coding assistant's capabilities and training data. Always review, test and validate AI-generated code before using it in production. The design files provide context, but the final code quality and correctness are the responsibility of the developer.

Approach comparison

Aspect	Managed Stacks	Custom Stacks
Code Generation	Template-based	AI-assisted
Technology Support	Predefined stacks	Any supported language
Pattern Flexibility	Fixed patterns	Custom patterns
Customization	Limited to stack options	Full customization
Implementation Speed	Immediate	Iterative

Both approaches ensure that the final implementation accurately represents the design specifications, though through different mechanisms.