BrighterSpace

Inspiration

Students at NYU use Brightspace for learning management every day. However, we noticed how fragmented the contents, grades, and other resources are in Brightspace. This motivated us to reimagine the learning management system with a new way for students to access their resources in a more integrated way in an AI-integrated, widget-based dashboard, making learning more accessible and engaging.

What it does

In BrighterSpace, we redesigned the dashboard by creating individual widgets for announcements, assignments, calendar, grades, and other resources, which expand into pop-up windows upon clicking. The key feature, AI integration, is accomplished through two modes: command mode and conversation mode. In command mode, the AI interprets structured commands that allow users to manipulate widgets in an integrated way, such as searching for assignments by a term and displaying their relevant grades. The conversation mode provides a standard chatbot and also helps with command syntax.

How we built it

Frontend - React / TailwindCSS

Backend - MongoDB, Express.JS

Challenges we ran into

One of the biggest challenges was designing a command language that was expressive enough to be useful while still being simple and intuitive for students. We had to ensure that commands could be chained safely and implement a confirmation checking option to avoid executing ambiguous actions. It was also challenging to unify data coming from different widgets into a clean, readable interface.

Accomplishments that we're proud of

We are proud of the integration between our widget-based dashboard and the AI-powered command mode, which forms the core of our system. The command mode in particular feels powerful yet intuitive, enabling students to accomplish multi-step tasks with a single, concise instruction. We are also proud of the widget system itself, which is clean, modular, and designed with extensibility in mind.

What we learned

Throughout this project, we learned how important interface design is when dealing with multiple sources of information, and also integrating it with AI in a responsible and interpretable way. Additionally, creating realistic mock APIs helped us think carefully about how real educational platforms structure their data and how decisions can be made to allow for better student accessibility.

What's next for BrighterSpace

Going forward, we want to integrate BrighterSpace with real Brightspace APIs so that students can use it with their actual courses. We also plan to introduce new widgets, such as tools for office hours, study groups, and attendance. Another goal is to allow users to create their own widgets from templates, making the LMS even more customizable and engaging.

Furthermore, we hope to expand the AI’s capabilities to support more complex chained commands, such as planning study schedules or summarizing announcements across multiple courses, while ensuring security and reliability.

Built With

• ai
• html
• javascript
• machine-learning
• mongodb
• react
• tailwindcss