Show the pipeline: session notes → learning parameters → content spec
This mirrors the original case study's three-prompt design: a session-analysis prompt that pulls learning signals out of the raw transcript, a content-generation prompt that designs a targeted practice problem around them, and a UI-generation prompt that turns that content into a component built from one shared set of design tokens.
Step 1: Session notes (from a session transcript)
...
Step 2: Learning parameters (JSON)
...
Step 3: Content spec (JSON)
...
Preview
Open in new tab ↗Nothing rendered yet. Pick an example above.