Smarter learning, with an app that turns handwritten notes into real exam questions
Today marks the launch of QueriU. We built this learning app together with heinrich+gretchen, from the first idea through to the finished product. It shows what happens when pedagogical expertise and technical execution are considered together from the start, rather than handled one after the other.
How QueriU works
The principle is simple. Children photograph their handwritten notes, and QueriU generates questions based on exactly the material they are currently studying. It is not an open chatbot that answers freely, but an app that guides students through their own study material with purpose. Passive re-reading becomes active recall, which is the learning method that research consistently shows works best.
The starting point was an observation, not a technology
In many families, parents take over quizzing their children before exams. They turn notebook entries into questions, listen to the answers, and try to gauge what has actually stuck. This costs evenings and nerves, and pedagogically it is often not the best approach anyway. This observation was the starting point for QueriU, long before any discussion of AI models or architecture decisions.
From there came a concrete question: could this be solved with AI, without children giving up the work of thinking for themselves?
The real challenge was pedagogical, not technical
Generating questions from text is something most language models can already do. The harder part is generating questions that genuinely require recall rather than simple recognition, questions that reveal the difference between "I have read this before" and "I can explain this." We spent considerable time refining that distinction, both in prompt design and in the technical architecture behind it, until the results felt right for the children using them.
Two teams, one shared responsibility
heinrich+gretchen was responsible for the product concept, design and UX/UI, while HipSquare handled the AI integration and technical architecture. What mattered most was that both perspectives were combined from the first sketch onward, rather than developed separately and merged at the end. In our view, that is what separates a technically impressive demo from a product children actually use.
This is generally how we approach projects like this one: working closely with the domain experts, taking clear technical ownership, and making sure the technology carries the concept rather than just following it.
Try QueriU
QueriU is live today at queriu.app. We look forward to hearing what you think.
If you're planning a similar project, want to bring an idea with AI support into a solid technical implementation, or are simply curious how a collaboration like this works with us, feel free to reach out to HipSquare.