The Coding5s Framework

Coding5s Architecture

The Framework

Coding5s is an open-source technical-learning framework designed for an environment where AI can generate answers almost instantly. Instead of removing AI from learning, the framework structures how learners practice, how AI assists them, and what context should remain available as the learning journey evolves.

Core principle: Learn With AI. Don’t Outsource Your Thinking.
PILLAR_01 / LEARNING

5-Stage Learning Architecture

Structures technical learning through five progressive forms of practice: Practice, Debug, Complete, Refactor, and Extend. The goal is to preserve meaningful learner participation while AI remains available as a tool.

What should the learner practice?
PILLAR_02 / BEHAVIOR

AI Mentor Swarm

Defines specialized AI behaviors and assistance boundaries. Mentors can explain, challenge, question, critique, or provide scaffolding without automatically performing the cognitive task the learner is expected to practice.

How should AI help?
PILLAR_03 / STATE

Stateful5s

Addresses cumulative learning by preserving selected technical state outside the LLM. Instead of depending entirely on conversation history, later work can receive the information it still needs from earlier lessons.

What must remain known later?
Structured Practice Pillar 1
Controlled AI Assistance Pillar 2
Cumulative Context Pillar 3
Scroll to Top