Design coherent courses from the papers and chapters you actually want to teach.
Upload your reading list. Get a prerequisite-aware schedule, anchored problem sets, and an agent that re-paces the course as your cohort progresses.
From a folder of PDFs to a 14-week course.
Upload your reading list
Drop in chapters and papers. Concepts get extracted and mapped to your outcomes.
Approve the outline
Review the week-by-week topic sequence before details generate. Drag, reorder, or ask the agent.
Generate, refine, re-pace
Per-week readings, discussion prompts, and a 10-question problem set. Re-pace mid-semester from cohort signal.
See real examples.
Two pre-built courses, ready to browse end-to-end.
Machine Learning Systems
Built from 16 papers on distributed training, parallelism, memory optimization, and LLM serving. Includes a concept dependency graph and 6 alternative learning-path orderings.
Calculus I
Built from an 8-chapter undergraduate calculus textbook. Limits, derivatives, integrals, differential equations, and series — sequenced with prerequisite-aware weekly readings.
What instructors are saying.
“I rebuild my reading list every fall. This is the first tool that actually understands the difference between a foundational paper and a survey, and orders them like a person would.”
“The problem set generator is the killer feature. Ten grounded MCQs with the source page anchored on each one. I edit two or three and ship it.”
“Re-pacing actually works. After the midterm I told the agent the cohort struggled with attention math; it pushed FlashAttn back a week and inserted a primer. Two minutes.”
Representative feedback from pilot users.