-
-
Primary CollabLens Dashboard
Inspiration
I started working with Dr. Steve Joordens who is a professor at the University of Toronto. Dr. Joordens runs multiple education and AI projects/startups. In collaboration with him I am working to build a student group work manager that helps inform professors on the group dynamics in his class.
What it does
On the platform professors can make a new assignment and upload an excel spreadsheet of student groups. This platform will then make a Discord server where each group has a private channel as well as a SharePoint folder for the group to work in. All of the group's messages from Discord as well as activity from SharePoint will be logged to a MySQL database. Gemini's API will use all of this data to generate insights that university professors can use to understand the dynamics of the team.
How we built it
The frontend is delivered via Apache web server and PHP (I am using UIkit for the UI's CSS). Data is stored in a MySQL database, and the Gemini API integration is done in the Node.js backend. The entire project's architecture is built using Docker.
Challenges we ran into
Difficulties with SharePoint and Docker API integration took too much time to much time to figure out how to implement so for the time being the project is using dummy data.
Accomplishments that we're proud of
I am very happy with the workflow that I have put together to provide valuable data driven insights.
What we learned
- Integrating OpenAI’s Gemini API to generate dynamic, is something I have only done once before. So I have learned a lot to use it to generate AI-powered group insights. ## What's next for CollabLens
- SharePoint API integration.
- Discord bot TA to help advise students based on their performance.
- Receive testing feedback from Dr. Joordens when he uses this platform in his Fall 2025 classes.
- Advanced data reports for researches interested in studying student group dynamics.
Log in or sign up for Devpost to join the conversation.