> "Let’s break down what it takes to create just one course – say, an introductory course on core pre-algebra concepts:<p>- 50+ core concepts to teach
- 20+ problems per concept
- → ~1,000+ individual problems per course<p>That last number kept us up at night. You can’t just explain a concept once and move on. You need enough variations to let learners truly master each idea, enough edge cases to build real understanding, and enough of a ramp in difficulty to create that perfect learning curve.<p>Designing the right game and sequence of concepts, so that learning feels like flow, is the fun part. But then you need to make a thousand carefully calibrated problems. And that part is a lot less fun – and it takes a long time."<p>This seems like a practical and appropriate use of LLMs. It reminds me of a similar application I heard about recently from a friend who teaches language classes using ChatGPT or equivalent to generate dozens of example sentences to teach specific grammar rules.<p>I find it refreshingly limited in scope compared to many projects that aim to outsource the creative process altogether rather than focusing on automating the legitimately menial and repetitive tasks.