QuizKar

Single-Page Quiz Architecture with AI-Powered Content

Overview

QuizKar is a high-performance assessment engine capable of generating localized quizzes instantly via Gemini AI. The system focuses on data-driven insights, providing real-time competitive analytics for students through optimized backend pipelines.

Key Features

  • AI Question Orchestration: Custom-built logic to parse Gemini API responses into validated JSON quiz structures.
  • Scalable Leaderboards: Real-time ranking calculated via MongoDB Aggregation Framework to ensure zero-lag standings.
  • Session Management: State-persistent quiz sessions using React hooks for a seamless educational experience.
  • REST API Architecture: Secured Node.js endpoints with robust error handling for quiz lifecycle management.

Technical Challenge

Scaling Real-Time Stats: Recalculating scores for thousands of students after every question can be DB-intensive.

Solution: I used MongoDB Aggregation pipelines to calculate scores only when requested, and explored caching strategies to minimize redundant calculations during live challenge modes.

The 60-Second Pitch

"QuizKar is a MERN-stack quiz platform I built that integrates AI for question generation. I used the Gemini API to let users create quizzes instantly from any topic. On the backend, I focused heavily on data efficiency—using MongoDB aggregations to handle real-time leaderboards. This project taught me how to handle asynchronous AI responses and manage high-frequency data updates."