-
-
configuration page of the quizes
-
interface to attempt the quiz
-
page to upload the contents to generae the quizes
-
Dashboard of the student based on their perfomance of the attempted quizes
-
teacher dashboard to create classroom and see the anlytics of classroom and students individually
-
student dashboard to generate quuzes and see their dashboard
-
ai tutor to solve the problem and doubt faced by the studnets during
-
quizes attempted by the studnets and suggestion by the ai
-
landing page
Inspiration
During my JEE preparation, I was overwhelmed by the large volume of study material such as PDFs, handwritten notes, coaching books, and YouTube lectures.
The biggest challenge was finding high-quality, topic-specific questions.
Searching for questions on a single topic often took hours, resources were scattered, and passive reading did not help with retention. Creating quizzes manually was nearly impossible under time pressure.
This led to a simple idea:
What if an AI could instantly read any study material and generate accurate, topic-wise quizzes and revision content?
That idea became QuizerAI — a platform that transforms any learning resource into personalized quizzes and summaries within seconds.
What It Does
QuizerAI is an AI-powered learning and assessment ecosystem that connects students, teachers, and institutions on one unified platform.
For Students
- Convert PDFs, images, handwritten notes, and YouTube videos into instant topic-wise quizzes
- Generate summaries, flashcards, and revision content
- Practice CBT-style exams for JEE, NEET, UPSC, CUET, SSC, and board exams
- Get instant explanations through a 24/7 AI Tutor
For Teachers
- Create virtual classrooms using join codes or QR codes
- Auto-generate quizzes, assignments, and homework from lesson content
- View real-time class analytics such as accuracy, weak topics, and completion rates
- Identify struggling students without manual effort
For Institutions
- Automated attendance through quiz participation
- Centralized dashboards for admins and departments
- Multi-class and multi-teacher management
How We Built It
Frontend
- Next.js, React.js, Tailwind CSS, Shadcn UI, TypeScript
Backend
- FastAPI (Python), Uvicorn/Gunicorn
- Asynchronous workers for large-scale quiz generation
AI & OCR
- AWS Bedrock (Claude, Llama)
- OpenAI and Groq inference
- Tesseract OCR and AWS Textract
- YouTube transcript extraction
Databases & Infrastructure
- MySQL (AWS RDS), Redis (ElastiCache)
- AWS S3, Lambda, EC2, Elastic Beanstalk
- CloudFront, Route53, ACM, Amplify
DevOps & Tools
- Docker, GitHub Actions
- PostHog analytics, Clerk/Auth.js
- CloudWatch monitoring
Challenges We Faced
- Achieving accurate OCR for handwritten notes using a Textract + Tesseract pipeline
- Managing AWS costs during sudden traffic spikes
- Handling YouTube API rate limits and incomplete transcripts
- Scaling FastAPI and Redis for thousands of concurrent users
- Designing a seamless UI/UX for students, teachers, and institutions
- Building a production platform alongside college academics
Accomplishments We’re Proud Of
- Built a complete AI-powered learning ecosystem
- Reached 10,000+ beta users organically
- Generated thousands of topic-wise quizzes automatically
- Implemented high-accuracy handwritten OCR
- Scaled to support thousands of concurrent users
- Reduced teacher workload through automation
- Deployed a secure and scalable AWS cloud infrastructure
What We Learned
- Active recall is more effective than passive learning
- Teachers need speed and simplicity, not complex tools
- AI works best when aligned with real classroom workflows
- Scalability, cost control, and UX are as important as model accuracy
What’s Next for QuizerAI
- Adaptive learning paths based on individual strengths and weaknesses
- AI-based exam proctoring for fair assessments
- Offline mode for low-connectivity regions
- Mobile app for students and teachers
- AI lecture notes generator from videos and handwritten pages
- Exam-specific modules for CBSE, ICSE, JEE, NEET, UPSC, SSC
- Predictive analytics to identify struggling students early
- LMS integrations with Google Classroom, Moodle, and Canvas
- Nationwide rollout across schools and colleges
- AI-moderated community doubt-solving spaces
- Teacher question-bank marketplace
Future Enhancements & Advanced AI Features
1. Smart Study Management (AI-Powered)
QuizerAI will introduce an intelligent study management system to help students plan, track, and optimize their academic workload using AI.
- Assignment Management: Track all assignments with subject, due date, estimated study hours, and difficulty level
- AI-Powered Scheduling (Planned): Automatically generate personalized study schedules by breaking large tasks into daily, manageable sessions
- Pomodoro Timer: Built-in focus timer linked to specific assignments with automatic progress updates
- Visual Analytics: Interactive charts showing weekly study hours, subject-wise distribution, and assignment completion status
- Smart Notifications: Automated reminders (3 days, 1 day, same day), daily study nudges, and inactivity alerts
- Gamification System: Achievement badges such as Early Bird, Night Owl, Perfect Week to encourage consistent study habits
- Export Functionality: Download professional PDF reports and CSV files for progress tracking and reviews
2. Voice-Based AI Tutor (Multimodal Interaction)
QuizerAI will enhance its AI Tutor with full voice and emotion-aware capabilities to support diverse learners.
- Multilingual Voice Support using SARVAM AI for speech-to-text and text-to-speech
- Emotion-Aware Learning using DeepFace for facial emotion analysis and OpenAI Whisper-v3 for speech emotion recognition
- Automated Reports generated after sessions, stored on IPFS, and delivered via Twilio WhatsApp API
- Enables more natural, accessible, and inclusive learning—especially for early learners and regional language users
3. Adaptive Testing & Personalized AI Tutor
QuizerAI will introduce an advanced adaptive learning engine that personalizes education for every student.
- Builds a personalized curriculum based on both selected course (Physics, Chemistry, Biology, Math, etc.) and future career goals (Doctor, Engineer, Researcher, Scientist, etc.)
- Uses an RNN-based memory model to track concept mastery and learning history over time
- Employs a Deep Reinforcement Learning teacher agent that continuously adjusts:
- Difficulty
- Pacing
- Concept order
- Difficulty
- All explanations, examples, and practice problems are generated via a RAG pipeline powered by katanemo/Arch-Router-1.5B to ensure grounded and accurate responses
- Learns the student’s learning style dynamically from mistakes, retries, and improvements
- Delivers a true one-on-one AI tutor experience, far beyond static courses or generic chatbots
PathLearn transforms QuizerAI from an assessment tool into a fully adaptive, career-aligned learning companion.
4. Exam-Level Adaptive Practice (School to Competitive Exams)
- Adaptive question generation from school-level to advanced competitive exams
- Difficulty automatically scales from fundamentals to exam-level problems
- Continuous alignment with exam patterns such as CBSE, ICSE, JEE, NEET, UPSC, SSC, and state boards
5. Collaborative & Competitive Learning
- Students can create group quizzes and compete with peers
- AI-based rankings, leaderboards, and performance comparison
- Encourages peer learning, healthy competition, and motivation
6. AI-Generated Teaching Content (For Schools & Teachers)
- Automatic generation of animated slides, PPTs, and visual explanations from uploaded content
- Makes classroom teaching more interactive and engaging
- Reduces teacher preparation time while improving concept clarity
These future features position QuizerAI as a complete, intelligent education ecosystem — covering learning, practice, evaluation, planning, and growth.
QuizerAI’s vision
Make learning active, fair, personalized, and accessible — so no student is left behind and no teacher is overburdened.
Built With
- amazon-web-services
- auth.js
- aws-acm
- aws-amplify
- aws-bedrock-(claude
- aws-cloudfront
- aws-cloudwatch
- aws-ec2
- aws-elastic-beanstalk
- aws-lambda
- aws-route53
- aws-textract
- clerk
- docker
- fastapi-(python)
- github
- github-actions
- groq-lpu-inference
- gunicorn
- llama)
- mysql-(aws-rds)
- next.js
- openai-apis
- posthog
- react.js
- redis-(aws-elasticache)
- shadcn-ui
- tailwind-css
- tesseract-ocr
- typescript
- uvicorn
Log in or sign up for Devpost to join the conversation.