Where Human Intelligence and Artificial Intelligence Collaborate in Software Development
Decision-Driven Software Engineering (DDSE) is a modern software development methodology that makes technical decisions first-class artifacts in the development lifecycle. DDSE bridges the gap between agile development practices and architectural governance, especially in the era of AI-assisted coding.
The foundational principle: "In the AI era, human value lies not in being the exclusive generator of solutions, but in being the authoritative decision-maker who guides generation toward human-aligned outcomes."
AI can generate code faster than humans—but it can't make intelligent decisions about what should be built or why. Without decision governance, AI-assisted development creates a dangerous gap: rapid implementation without human authority over architectural direction.
The Critical Problems:
- AI generates code without understanding your architectural decisions and constraints
- Teams lose decision context as AI accelerates development cycles
- Technical debt compounds as AI suggestions conflict with undocumented design principles
- Human decision authority gets eroded by the speed of AI implementation
The DDSE Solution: DDSE preserves human decision authority while unleashing AI implementation power. By making technical decisions explicit and structured, you create "decision memory" that:
✅ Guides AI tools with your authorized architectural constraints
✅ Preserves human authority over system evolution and direction
✅ Accelerates AI effectiveness through rich decision context
✅ Prevents architectural drift caused by uninformed AI suggestions
The bottom line: Teams using AI without decision governance are building fast toward chaos. DDSE ensures you harness AI's speed while maintaining human-directed, intentional architecture.
Make your decisions AI-ready before AI makes decisions for you.
- New to DDSE? Start with the DDSE Manifesto and Core Principles
- Academic Context? Read our Research Preprint for theoretical foundation
- Want to adopt DDSE? Follow the Implementation Guide
- Looking for templates? Explore TDR Templates with AI context support
- Integrating with Agile? See our Agile Integration Guide
- Need validation tools? Use our TDR Validator and ANTLR Grammar
ddse-foundation/
├── manifesto/ # Core philosophy and principles
│ ├── ddse-manifesto.md # Main manifesto document
│ └── author-bio.md # Founding researcher biography
├── principle.md # Philosophical foundation document
├── PREPRINT.md # Academic research preprint
├── ddse-spec-v1.0.md # Complete DDSE specification
├── tdr-templates/ # Standard TDR templates with AI context
│ ├── mdd-template.md # Major Design Decision template
│ ├── adr-template.md # Architectural Decision Record template
│ ├── edr-template.md # Engineering Decision Record template
│ ├── idr-template.md # Implementation Decision Record template
│ ├── tdm-template.md # Technical Decision Memo template
│ └── README.md # Template usage guide
├── adoption/ # Implementation guidance
│ └── implementation-guide.md # Flexible adoption strategies
├── integration/ # Framework integration guides
│ ├── agile-integration.md # Natural Agile enhancement patterns
│ └── README.md # Integration overview
├── tools/ # Validation and development tools
│ ├── TDRGrammar.g4 # ANTLR4 grammar for TDR parsing
│ ├── tdr_validator.py # Python TDR compliance validator
│ └── README.md # Tool usage documentation
├── community/ # Community governance and participation
│ ├── governance.md # Foundation governance structure
│ └── README.md # Community hub navigation
└── images/ # Documentation assets and diagrams
DDSE introduces Technical Decision Records (TDRs) as structured artifacts that capture technical choices with AI-ready context:
- MDD (Major Design Decision): Strategic architectural and technology choices
- ADR (Architectural Decision Record): System architecture and design decisions
- EDR (Engineering Decision Record): Development practices and tool decisions
- IDR (Implementation Decision Record): Component and code-level decisions
- TDM (Technical Decision Memo): Lightweight decision documentation
All TDR templates include:
- YAML frontmatter for structured metadata
- AI context sections for intelligent tool integration
- Cross-reference support for decision traceability
- Validation-ready structure for automated compliance checking
- 🎯 Reduced Architectural Drift: Explicit decisions prevent uncontrolled system evolution
- 🤖 AI-Enhanced Development: AI tools work intelligently within documented decision context
- 📚 Preserved Decision Knowledge: Rationale and trade-offs are captured for future reference
- 🔍 Enhanced Traceability: Clear links between decisions, implementation, and outcomes
- ⚡ Faster Team Onboarding: New members understand system evolution and constraints
- 🔧 Automated Validation: Tools ensure decision documentation compliance and quality
- 🧠 Human-AI Collaboration: Preserve human decision authority while leveraging AI capabilities
- Understand the Philosophy: Read the Core Principles and Manifesto
- Explore the Specification: Review the DDSE Specification for complete methodology
- Try the Templates: Use our TDR Templates for your first decisions
- Validate Your Work: Run the TDR Validator to ensure compliance
- Integrate Naturally: Follow our Agile Integration Guide for seamless adoption
- Join the Community: Connect through our Community Hub
DDSE includes comprehensive tooling for automated validation and compliance:
-
TDR Validator: Python-based validation engine
- Validates YAML frontmatter and required sections
- Checks cross-references and AI context sections
- Supports all TDR types with type-specific rules
- CLI interface with text and JSON output formats
-
ANTLR4 Grammar: Complete grammar specification
- Defines TDR markdown structure for parsing
- Enables language-agnostic validator implementations
- Supports frontmatter, sections, and cross-references
Usage Examples:
# Validate a single TDR file
python tools/tdr_validator.py tdr-templates/adr-template.md
# Validate entire directory with strict mode
python tools/tdr_validator.py tdr-templates/ --strict
# Generate JSON validation report
python tools/tdr_validator.py tdr-templates/ --output jsonThe DDSE Foundation operates through collaborative governance with multiple ways to participate:
Leadership Structure:
- Founding Chair: Mahmudur Rahman Manna - Convener, Researcher, and Principal Architect
- Community Stewards: Open positions - apply through governance process
- Executive Committee: Technical Director, Community Manager, Industry Liaison (positions open)
- Advisory Board: Academic Advisor, AI Ethics Advisor, Enterprise Architect (positions open)
- Working Groups: Specification, Tooling, and Education leads (positions open)
Get Involved:
- Review our Governance Structure for leadership opportunities
- Join Community Discussions through our communication channels
- Contribute to Tools and Documentation via GitHub
- Share Your DDSE Experience through case studies and feedback
Contact:
- Foundation Repository: GitHub
- Community Hub: Community README
- Founding Chair: Mahmudur Rahman Manna (mahmudur.r.manna@gmail.com)
- Leadership Contact: Full governance information
This project is licensed under the MIT License - see the LICENSE file for details.
If you use DDSE in your research or practice, please cite our foundational work:
@misc{ddse2025,
title={Decision-Driven Software Engineering (DDSE): Preserving Human Agency in AI-Assisted Development},
author={DDSE Foundation},
year={2025},
howpublished={\url{https://github.com/ddse-foundation/ddse-foundation}},
note={Academic preprint available at PREPRINT.md}
}For academic contexts, also reference our research preprint: PREPRINT.md
Join the DDSE Foundation • Community Hub • Governance • Academic Research