User Research

We conducted one interview with one high school student (Andie) and one agent, who we named Emily (Navigator🐊).

Students understand science better when they can test it by making something that works or fails. 

AI is most effective when it helps students translate their ideas into games rather than doing the thinking for them.

 Learning engagement drops when activities feel like “educational games” instead of real games.

Students prefer clear concept grounding before game creation.

Students are Struggling with Math

High school students often struggle to engage with mathematical modeling tasks (particularly when asked to interpret real-world situations and decide how to represent them mathematically).

Lack of Real Life Relevance

Students struggle to initiate mathematical modeling because they lack opportunities to practice translating real-world problems into mathematical representations (Krutikhina et al., 2018).

Problem

Feature title.

Feature description.

Feature title.

Feature description.

Research

Meet Our Users

These personas represent two learners who engage with science through game creation, ranging from a student already motivated by games to another who struggles with traditional science instruction.

Storyboards

The storyboard illustrates Andrew’s end-to-end learning journey, showing how structured stages, AI support, and reflection help him translate scientific concepts into a playable game and deeper conceptual understanding.