Imagine a Grade 4 teacher in Ghana asks an LLM to make a week’s multiplication lesson. The model can produce a polished teacher guide, worksheet, and practice game in seconds. The problem is that the material can look completely reasonable while being ahead of what the local curriculum expects at that grade.
That is a hard problem to solve with prompt engineering alone. The model needs curriculum context: not just mathematics, Grade 4, but the exact expectations for that curriculum, how they sit within the source document, what smaller skills they contain, what earlier or related learning connects to them, and where those claims came from.
A plausible lesson is not necessarily a curriculum-grounded lesson. The application needs to know what the local curriculum actually expects. Curriculum PDFs already contain much of that context, but PDFs are a poor interface for traditional software. Our answer is to turn them into a structured knowledge layer that keeps the source curriculum intact, adds useful graph relationships around it, and makes the result available to AI applications through a common interface.
Our approach: we reconstruct the PDF before interpreting it, preserve the curriculum’s own hierarchy, break broad statements into smaller pieces of learning, add reviewed progression relationships, and serve the resulting graphs through MCP.
Usually, I’d say the question doesn’t even matter. But here’s one that is topical:
“How do you measure impact when the treatment itself is a moving target, evolving alongside the control?”
This comes up constantly in AI evaluation conversations. Traditionally, A/B testing assumes your treatment is consistent across the population. But good tech solutions, and AI products in particular, are constantly evolving. You are (and should be) always experimenting to improve your product.
Creating good surveys is hard. Ask any economist or researcher who’s spent weeks crafting the perfect questionnaire, and they’ll tell you about the countless hours spent hunting through academic papers, existing surveys, and their own memory banks, searching for question formats, flow structures, and methodological approaches that can help them build a cohesive survey. Even when they need to develop original questions, they’re often looking for proven frameworks and industry-standard approaches that can serve as reliable benchmarks for their own work.
SurveyStream is a software product developed by IDinsight to support and streamline primary data collection operations. SurveyStream helps data collection teams manage survey operations more efficiently, freeing up their time to focus on other crucial activities both before and during a survey.
SurveyStream functions as a web application backed by a robust data management system that survey teams can use to manage enumerator hiring, assign enumerators to respondents, send periodic emails to enumerators with their assignments and get timely insights into productivity and data quality.
AI has the power to free up human time and improve lives. Yet it can only do so if it is designed ethically and inclusively. There’s a significant risk that AI will primarily extract from, rather than serve, ordinary people. Just as with the industrial and digital revolutions before, people may become further alienated from the world around them. We want to forestall this, without adding drag to the full and remarkable potential for social impact from technology.
We propose two new principles to drive AI development toward promoting dignity and human connection, ensuring that users - students, patients, citizens - are treated with respect for their humanity. In this fast-moving field we hope these will guide IDinsight’s own work, and we welcome feedback on how to improve them further.
