Get to know us
About Documentation Archivist™
AI is powerful. But without structure, it forgets.
Documentation Archivist was created to solve a simple problem: how to preserve context across AI sessions without retraining models, losing control of your work, or repeatedly reconstructing the same explanations.
It introduces a bounded, inspectable continuity layer for serious AI use. Instead of relying on a model’s internal memory, Documentation Archivist stores structured artifacts outside the model—so work can be retrieved intentionally rather than regenerated repeatedly.
The goal is not to replace AI. It is to stabilize it.
By favoring retrieval over recomputation whenever possible, Documentation Archivist is designed to reduce unnecessary processing and the operational waste associated with repeated AI use. Structure improves not only clarity, but efficiency.
This approach is AI-agnostic and portable. It is intended for professionals who need continuity across days, weeks, or months of work—without sacrificing transparency or control.
Documentation Archivist was developed by Marcia L. Coulter through independent research into AI continuity, provenance, and bounded memory systems. It reflects a belief that AI works best when it operates within clear, inspectable constraints.
AI needs structure. Documentation Archivist provides it.
Marcia L. Coulter
Founder
For much of my career, I’ve worked at the intersection of documentation, learning systems, and technical workflows. In very different environments — instructional design, federal contracting, data quality review, consulting — I kept encountering the same pattern:
Organizations lose continuity.
Between drafts.
Between semesters.
Between tools.
Between people.
Important reasoning disappears. Work gets repeated. Errors compound quietly.
Documentation Archivist™ grew out of a simple question:
What if documentation were treated not as paperwork — but as preserved thinking?
DA is built on the principle of bounded, inspectable continuity. It separates reasoning from output, preserves intellectual lineage, and supports responsible AI use without surrendering ownership.
I did not set out to build a software product. I set out to solve a recurring structural problem — one I had seen across disciplines and institutions. The work reflects a long-standing commitment to clarity, responsibility, and systems that help people think more effectively together.
Kyoto Moon exists to build knowledge tools with roots — tools designed to endure, not just to accelerate.