Make the Work Visible
Separate the responsibilities that govern a project, control its research, preserve its accepted knowledge, and direct its execution.
About DCPF
DCPF is an independent experimental project focused on organizing sustained human-AI collaboration into a visible, reviewable, and maintainable working system.
The Project
The Distributed Cognitive Prompting Framework for AI Collaboration, or DCPF, organizes collections of AI interactions through four connected frameworks: Governance, Research, Knowledge, and Runtime.
The project explores a practical question: how can people move beyond isolated prompts and preserve the rules, evidence, accepted knowledge, task instructions, human decisions, and review processes needed for sustained work with AI?
Prompt engineering can improve an individual interaction. DCPF focuses on organizing collections of interactions into a persistent architectural system.
Purpose
Keep Humans in Authority
Make human judgment, verification, acceptance, revision, rejection, and accountability explicit throughout the collaboration.
Support Continued Work
Preserve the information and operating instructions that allow important AI-assisted work to continue across sessions, tools, and contributors.
Enable Responsible Testing
Provide a shared structure that people can examine, adapt, test, critique, and validate in their own domains.
Creator and Author
DCPF was created and is authored by Ralph A. Perez, Jr. The framework grew from practical experience organizing complex, continuing work with AI and from the effort to make that collaboration easier to inspect, maintain, and teach.
The official website, Introductory Edition, framework explanations, starter materials, and continuing project updates are developed as parts of the same independent DCPF project.
Experimental Status
DCPF has demonstrated promise through its development work, practical applications, and training practicums, but it has not been broadly or independently validated across domains, organizations, AI systems, or production environments.
Users should begin in controlled, low-risk testing environments, validate the approach for their own domain and tools, and retain human judgment and accountability. DCPF should not be presented as a proven universal method or as a substitute for professional review.
An Independent Framework
DCPF is not an AI model, chatbot, software platform, certification program, or professional service. It is an organizational framework that people can use to structure their own AI-assisted work.
The project is not part of, sponsored by, endorsed by, or affiliated with any AI provider merely because its materials discuss or can be used with that provider’s systems. Product names, company names, and trademarks belong to their respective owners.
Free Official Materials
DCPF learning materials are being made freely available through this website, including the framework pages, implementation guidance, starter artifacts, and the forthcoming reader-ready Introductory Edition.
dcpfai.org is the authoritative site and definitive reference for current official DCPF materials, editions, permissions information, and updates.