Four lessons, one per screen. The 4D framework, put to work on course design, materials and assessment. Each slide is a thing to point at while you explain. Take the silences.
Two moves: get fluent in the 4Ds, then put them to work: first designing the course, then building what students actually use.
AI gets useful when it knows your situation. Spend the first session turning your teaching into one reusable document.
Ten minutes of self-reflection first (values, constraints, who your students are, the methods you trust), then let Claude interview you and synthesise it into one document. Every later conversation starts already onboarded.
Augmentation, not automation: you stay in the work. Four habits, each with three parts, and a loop in the middle that turns a tool into a thinking partner.
Describe, then discern, then describe again: that continuous loop is the engine. Delegation sets it up; Diligence keeps you responsible for accuracy, bias and transparency. (Full breakdown: Anthropic's AI Fluency: Framework & Foundations.)
Plan, set the context, then work the loop across three design tasks, and document what you decided and why.
For every suggestion, ask "does this serve my students?" and explain your reasoning back: build on good ideas, redirect weak ones. Ask the AI to take the students' view at transition points: what might confuse them, what scaffolding is missing.
Same three steps for every material: set the task on top of your context, refine in the loop, then run the Diligence check before it reaches a student.
Established context is what makes materials cohere: slides, guide and quiz line up because they share one source. Academic integrity cuts both ways: when students have AI too, fluency helps you design assessments that ask for authentic learning.
Build your teaching context, run the 4Ds, and let the describe–discern loop make AI a thinking partner for design, materials and assessment.
tutto.one/praxis · daniel@tutto.one