Access our app here: http://143.198.43.88/ Watch a video Demo here: https://youtu.be/gWMAToZywtI
- Hardik Narkhede - hardik.narkhede@mail.utoronto.ca
- Vraj Shah - vrajr.shah@mail.utoronto.ca
- Arhum Parvez - arhum.parvez@mail.utoronto.ca
Open Quizzer is an online quizzing platform designed to go beyond traditional multiple-choice apps like Kahoot by using open-ended questions and real-time Large Language Model (LLM) summarization.
The goal is to make classroom interaction more meaningful, inclusive, and engaging by allowing students to submit free-text or code-based responses — which are then categorized and summarized live using AI.
-
Open-ended Response Collection
Unlike Kahoot, Open Quizzer allows quiz respondents to submit short or long-form textual answers. -
Real-time LLM Summarization
An integrated LLM (like OpenAI GPT) will instantly summarize classroom responses into key topics or themes, similar to YouTube's comment summarization. -
Interactive Coding Questions (Future thing, last sprint)
Students can submit code snippets, which the LLM will analyze for correctness, common mistakes (e.g., missing semicolons, wrong order of operations), or according to instructor-defined rules. -
Passive-to-Active Learning
Encourages participation by requiring original answers, helping passive students become active contributors. -
Instructor Dashboard
Real-time dashboard for the Quiz Master to view themes and trends across student answers.
- Our frontend framework of choice will be AngularJS
A critical part of this project is real-time interactivity. The app will reflect user changes immediately without requiring manual refreshes using:
- WebSockets (e.g., Socket.io) for real-time communication
- LLM prompt processing and updating the frontend
- Live display of categorized answers as students submit them
- Quiz Master dashboard for analyzing trends/gaps
- Quiz Master can open a poll question
- Quiz Respondents can submit text-based answers
- LLM summarizes common topics/themes from responses
- OAuth 2.0-based authentication
- Stripe integration
- Quiz Master dashboard with real-time visual summaries
-
Code snippet response type
Students submit language-specific code responses, and the LLM:- Highlights common syntax/logic issues
- Flags critical errors (e.g., missing
await, incorrect variable casing) - Marks according to instructor-defined rule sets (e.g., camelCase enforcement)
-
Participation Tracking
Logs all respondents per quiz for participation grading -
QR Code Entry
Join quizzes quickly by scanning a QR code instead of a unique code
Full Tech Stack: AngularJS, Express.js, PostgreSQL, Socket.io, OpenAI API, Stripe