Inspiration
I've always wanted to make an impact in education, as I see education as the gateway to opportunities. I believe that empowering students in education will shape a better future for not only themselves, but for generations to come as the youths of today are the leaders of tomorrow.
What it does
Our project focuses on adapting tests based on student's weaknesses and strengths to help them push their limits and promote continuous improvement. Furthermore, we understand that teachers and schools lack the data needed to provide targeted help to students which is why we decided to introduce automated grading in order to track student progress and provide schools and teachers with actionable data.
How we built it
We combined manual development with Vibe Coding for rapid frontend iteration. Using VS Code and the Cline extension, we integrated Gemini 2.5 Pro directly into our IDE for AI-powered co-development, allowing faster prototyping and debugging. The backend leveraged Google Firebase for user authentication and real-time data storage—chosen specifically for its scalability, reliability, and zero-cost tier during development. This helped us focus more on feature implementation and less on infrastructure.
Challenges we ran into
One of the major challenges was implementing a dynamic data model to support multiple subjects and associated test scores for individual students. Firebase’s NoSQL structure introduced some quirks in retrieving and updating nested data efficiently. Debugging read/write errors during early database integration was especially time-consuming and required close inspection of asynchronous data handling and document referencing.
Accomplishments that we're proud of
A key technical highlight was integrating a large language model (LLM) to automatically classify student quiz questions into difficulty levels (easy, medium, hard). This added a layer of intelligence to the system and removed the need for manual tagging—greatly enhancing personalization and adaptive learning.
What we learned
Using Vibe Coding significantly accelerated frontend development, and having Gemini 2.5 Pro embedded directly in our editor proved to be a game-changer for productivity. We gained practical experience setting up Google Authentication and appreciated how accessible and fast the implementation was. This project also reinforced how frontend development, especially with frameworks like Tailwind and Next.js, can be more streamlined compared to backend logic when equipped with the right tools.
What's next for EduSync
We plan on conducting market research via interviews with our potential users and participate in a few other pitch competitions to further validate our idea before launching our pilot in tutoring centers around New South Wales. We plan on getting feedback and improving our MVP using the build-measure-learn cycle to build the perfect EdTech Solution.
Log in or sign up for Devpost to join the conversation.