Engineering Knowledge Framework is an open-source framework for building AI-native engineering knowledge systems.
It provides a structured foundation for capturing, organizing, maintaining, evolving and applying engineering knowledge in a way that benefits both engineers and AI systems.
This project is not simply a documentation repository.
It is a framework for engineering knowledge.
Engineering organizations accumulate valuable knowledge every day through projects, successes, failures, architectural decisions and operational experience.
Unfortunately, much of this knowledge remains fragmented, undocumented, or is lost as teams evolve.
The Engineering Knowledge Framework exists to transform that experience into reusable organizational knowledge.
Our long-term vision is to establish an open standard for engineering knowledge management that enables organizations to:
- Capture institutional knowledge.
- Improve engineering decision making.
- Standardize engineering practices.
- Accelerate onboarding.
- Reduce repeated mistakes.
- Build engineering capabilities.
- Integrate AI throughout the engineering lifecycle.
- Continuously improve through organizational learning.
Engineering knowledge should scale with both people and AI.
The framework is built around several core principles.
| Principle | Description |
|---|---|
| Knowledge over Documentation | Capture understanding, not just information |
| Principles over Tools | Prefer timeless engineering principles over technologies |
| Explain Why | Document reasoning, not just procedures |
| Modular Knowledge | Compose knowledge from focused, reusable documents |
| Capability-driven | Organize around engineering capabilities |
| AI-native | Treat AI as a first-class consumer of knowledge |
| Continuous Improvement | Evolve knowledge through organizational learning |
These principles guide every aspect of the repository.
The framework consists of several complementary components.
Defines how engineering knowledge is created and maintained.
Examples include:
- Repository philosophy
- Writing standards
- Contribution guidelines
- Knowledge architecture
- Document taxonomy
Each engineering discipline is represented as a capability containing all knowledge required to understand and execute that discipline.
A capability typically contains:
- Handbook
- Guides
- Playbooks
- Checklists
- Templates
- References
- Glossary
- Learning Paths
- AI Workflows
Examples include:
- Rails Engineering
- Software Architecture
- Security Engineering
- Testing
- Accessibility
- Platform Engineering
The framework treats AI as an engineering collaborator rather than simply a documentation generator.
AI is integrated throughout the engineering lifecycle, including:
- Context engineering
- Architecture reviews
- Code reviews
- Documentation generation
- Test generation
- Security analysis
- Accessibility assessment
- Refactoring
- Knowledge extraction
Human engineering judgement remains the final authority.
Engineering knowledge should enable continuous learning.
Capabilities may include:
- Learning paths
- Recommended reading
- Exercises
- Competency progression
- Practical examples
The framework is designed to support automation.
Examples include:
-
Repository generators
-
Templates
-
Linters
-
Link validation
-
Style validation
-
Knowledge graph generation
-
AI context generation
-
Search and indexing
-
Repository Organization
The repository is organized into several major areas.
governance/
Repository philosophy, standards and guidance
glossary/
Canonical engineering terminology
handbook/
Engineering principles and standards
guides/
Focused explanations of engineering topics
playbooks/
Repeatable engineering workflows
checklists/
Verification and quality gates
templates/
Reusable engineering artifacts
references/
Quick reference material
learning-paths/
Structured learning journeys
assets/
Shared diagrams, images and supporting resources
Each document type has a clearly defined purpose.
See:
for more information.
Engineering knowledge evolves continuously.
Experience
↓
Capture
↓
Review
↓
Refine
↓
Organize
↓
Publish
↓
Apply
↓
Learn
↓
Improve
↓
Repeat
This framework exists to support every stage of that lifecycle.
Traditional documentation is written solely for humans.
The Engineering Knowledge Framework is designed for both humans and AI.
Documents should therefore:
- Explain reasoning.
- Use consistent terminology.
- Cross-reference related knowledge.
- Separate principles from implementation details.
- Be modular and reusable.
- Be easy to navigate programmatically.
Well-structured knowledge enables AI systems to provide more accurate, consistent and context-aware assistance.
The framework is intended for:
- Software Engineers
- Senior Engineers
- Staff Engineers
- Architects
- Engineering Managers
- Technical Leads
- Platform Engineers
- Quality Engineers
- Open-source contributors
Organizations may adopt the framework in its entirety or adapt individual components to meet their own needs.
If you are new to the repository, begin with the governance documents.
These documents explain the philosophy, organization and contribution model of the framework.
Once you understand the philosophy, use the Table of Contents to explore the repository by capability or by topic.
The TOC provides two complementary views:
- From Top to Bottom: Every document organized by capability (Engineering, Architecture, Rails, Security) and document type.
- Cross-Cutting Topics: Documents grouped by topic (Testing, Security, API Design, Performance, etc.) across capabilities.
Each capability handbook also includes a Capability Map section listing all documents within that capability.
The current priorities are:
- Establish the governance framework.
- Build the foundational engineering capabilities.
- Develop Rails Engineering as the reference capability.
- Develop AI Engineering practices.
- Create reusable templates and tooling.
Progress is tracked in:
The Engineering Knowledge Framework aims to become:
- A reusable framework for engineering organizations.
- A reference implementation for engineering knowledge management.
- A foundation for AI-assisted engineering.
- A community-driven collection of engineering best practices.
- An open ecosystem for continuously evolving engineering knowledge.
The framework is intended to be adopted, extended and improved by the engineering community.
Contributions are welcome.
Before contributing:
- Read the governance documents.
- Understand the document taxonomy.
- Follow the Style Guide.
- Improve existing knowledge before creating new documents.
- Keep documentation timeless, modular and well connected.
See:
for the complete contribution workflow.
This repository is open source.
Unless otherwise specified, contributions are licensed under the repository's license.