ExamiQ

Inspiration

Preparing for competitive STEM exams and CBSE Class 11/12 boards often feels like drowning in heavy PDFs. We noticed that students lose dozens of hours simply context-switching between 100-page NCERT chapter textbooks, scattered Past Year Question (PYQ) banks, and handwritten formula sheets. Existing online EdTech platforms are bloated, require constant high-speed internet, and introduce endless distractions. We wanted to build ExamiQ—a clean, privacy-first, 100% offline study intelligence tool that runs locally on any student’s laptop without depending on external cloud servers or active Wi-Fi.

What it does

ExamiQ transforms static NCERT textbook PDFs and PYQ documents into an interactive, zero-latency learning engine directly on your device:

Offline PDF & PYQ Parsing: Ingests raw chapter PDFs and automatically categorizes past exam questions by exact sub-topics and mark weightage.

Instant Formula & Flashcard Extraction: Automatically isolates core mathematical equations, physical constants, and key definitions into searchable quick-reference cheat sheets.

Local Mock Quiz Engine: Dynamically generates custom practice tests filtered by difficulty, allowing students to test themselves offline with instant local scoring.

Zero-Latency Search: Lets students query formulas or key concepts across all stored chapters in under 10 milliseconds.

How we built it

Backend: Built using a lightweight Python architecture for fast local PDF extraction, text tokenization, and vector indexing stored locally on disk.

Frontend UI: Built with an offline-cached HTML/CSS/JS interface connected via a high-speed local WebSocket/Flask bridge for zero load lag.

Data Processing: Implemented modular Python parsing algorithms to extract formulas, LaTeX mathematical structures, and topic headers from unstructured textbook pages.

Challenges we ran into

Handling Complex Mathematical Formatting: Parsing inline formulas, scientific notation, and multi-line equations directly from PDF structures without losing spatial context or mathematical symbols.

Strict Offline Constraint: Ensuring all indexing, searching, and quiz-generation logic runs smoothly on local device hardware without relying on heavy cloud LLM endpoints or external APIs.

Zero-Latency Querying: Optimizing disk storage and local memory caching so searching across multiple heavy textbooks returns instant results.

Accomplishments that we're proud of

Designed a completely 100% local, privacy-first system that requires zero internet connectivity once installed.

Built a seamless parsing pipeline that turns unorganized PYQ question banks into indexed, topic-wise flashcards in seconds.

Created a lightweight, distraction-free user interface focused purely on high-speed execution and student productivity.

What we learned

Deepened our understanding of local PDF parsing, regular expression pattern matching, and structuring unstructured text data in Python.

Learned how to build robust, offline-first web interfaces that maintain persistent state and zero-latency performance using client-side caching.

Experienced the importance of designing software around real-world constraints—like spotty connectivity and time-sensitive student workflows.

What's next for ExamiQ

Multi-Subject Expansion: Broadening automated parsing pipelines to support Biology diagrams, Computer Science code snippets, and Chemistry reaction mechanisms.

Spaced Repetition System (SRS): Integrating an automated review algorithm (similar to Anki) that re-tests students on weak formulas based on their past quiz scores.

Desktop Bundle: Packaging ExamiQ into a single executable installer (PyInstaller/Electron) for one-click setup on Windows, Mac, and Linux.

Built With

• css3
• flask
• html5
• javascript-(es6+)
• pypdf
• python
• websockets