Inspiration
Students don't all learn the same way, yet most AI study tools give everyone essentially the same explanation. We wanted to build a study companion that adapts to the learner instead of forcing the learner to adapt to the AI.
What it does
Étudier is an adaptive AI study copilot that changes how it explains concepts based on how a student learns best. It can turn difficult topics into simple explanations, step-by-step breakdowns, analogies, examples, stories, bullet points, or visual summaries.
How we built it
We built Étudier around a personalized learning pipeline:
User Question → Learning Profile → Adaptive Prompting → AI Engine → Interactive Workspace
Students can provide topics, questions, notes, or reference material, while the system uses their learning preferences and academic level to shape the response.
Challenges we ran into
The biggest challenge was making personalization actually useful instead of simply changing the wording of an answer. We had to design a system that could consistently control explanation depth, tone, structure, examples, and complexity while keeping the underlying information accurate and useful.
Accomplishments that we're proud of
We're proud of building a study experience that treats personalization as a core part of the product rather than an optional AI setting. Our adaptive explanations, ELI5 mode, grade-level scaling, and interactive whiteboard demonstrate how AI can make learning more approachable and less overwhelming.
What we learned
We learned that good educational AI isn't just about generating better answers. It's about understanding how the answer should be delivered. Personalization, cognitive load, clarity, and interaction can be just as important as the underlying AI model.
What's next for Étudier
We want to make Étudier increasingly personalized by learning from student interactions, improving its learning-style detection, expanding the interactive whiteboard, and supporting richer study materials such as PDFs, notes, diagrams, and course content.
Built With
- gemini
- google-cloud
- groq
- gsap
- javascript
- nextjs
- sass

Log in or sign up for Devpost to join the conversation.