An AI-native engineering knowledge repository that captures decisions once and derives documentation, architecture, and technical content from a single source of truth.
Engineering decisions, knowledge capture, and continuous learning while building software in public.
This repository documents the engineering journey behind building Creator Labs AI.
Unlike traditional documentation that focuses primarily on features or APIs, this repository captures the thinking behind engineering decisions:
- Why a decision was made
- Alternatives that were evaluated
- Trade-offs that were accepted
- Lessons learned
- Long-term architectural evolution
The goal is to create a permanent engineering knowledge base that benefits both humans and AI.
engineering-journal/
│
├── Engineering-Knowledge-Record(EKR)/
│ ├── templates/
│ └── Phase-*/
│
├── Journal/
│ └── 2026/
│
└── README.md
Engineering Knowledge Records capture significant engineering decisions.
Each EKR records:
- Context
- Problem Statement
- Options Considered
- Decision
- Trade-offs
- Lessons Learned
- Future Considerations
- AI Metadata
- Content Distribution
EKRs represent the canonical engineering knowledge for the project.
The Journal records the engineering journey as it unfolds.
Examples include:
- Human ↔ AI engineering conversations
- Sprint retrospectives
- Design discussions
- Engineering experiments
- Development notes
- Weekly reflections
Unlike EKRs, journal entries are chronological and exploratory.
The repository follows a few guiding principles:
- Capture knowledge once.
- Reuse knowledge everywhere.
- Automate repetitive engineering work.
- Prefer capabilities over pages.
- Prefer lean development over premature optimization.
- Record decisions, not just outcomes.
- Make documentation AI-friendly by design.
Engineering Work
│
▼
Journal Entry
│
▼
Engineering Knowledge Record (EKR)
│
├── Architecture Documentation
├── ADRs
├── Engineering Blogs
├── LinkedIn Articles
├── Reddit Discussions
├── Release Notes
└── Future AI Workflows
The Journal captures what happened.
The EKR captures what matters.
The repository evolves alongside the product.
Each engineering milestone is tagged to preserve:
- Engineering knowledge
- Decision history
- Documentation state
- Architecture evolution
MIT License