Design Concepts

System design documentation for AgentWay's gamification and learning experience.

6-stage progression based on Amazon's engineering ladder. Meaningful stages over arbitrary numbers.

Daily caps, depth multipliers, and sustainable progression. Quality over quantity.

Three achievement tracks: Knowledge, Practice, Habit. Clear progression paths.

Critical assumptions, Hooked model, SDT, JTBD. Frameworks for product-market fit.

Investment, Relatedness, Flow, Sharing. Actionable strategies for engagement.

3 Tracks, 12 Paths, ~40 hours. Frequent milestones + branching specialization = sustained engagement.

Claude Agent SDK as practice substrate. 13 APIs mapped across 12 Paths with progressive complexity.

Concept Clarification

AgentWay's official community bot. No deployment needed — join our Discord and use learning commands directly.

Open-source self-hosted AI assistant. Users deploy their own instance, sync learning data via AgentWay OpenAPI + token.

Design Principles

Information when needed, invisible when not.

Stages map to real-world competency.

Daily caps over unlimited grinding.

Quality engagement matters most.