GRIT — AI-Powered Exam Prep Platform

GRIT is a full-stack exam preparation platform for competitive exams in India — currently covering 14 GATE engineering branches and the CAT MBA entrance exam — with AI-driven personalization: weak-topic detection, adaptive mock test generation, and goal-gap analysis.

Features

• Multi-exam support — GATE (CSE, ECE, EE, ME, CE, AE, IN, CH, BT, MT, PH, AR, CY, DA) and CAT (VARC, DILR, QA), with a schema designed to extend to more exams.
• Real question corpus — hundreds of GATE questions (2020–2026) and 700+ CAT questions (2021–2024) parsed directly from official/source PDFs into a structured database, with topic classification.
• Adaptive mock tests — generate a personalized mock test weighted toward a student's weak and medium topics, with adjustable difficulty and length.
• Full test-taking flow — timed attempts, question navigator, mark-for-review, save/resume, auto-submit on timeout, MCQ/MSQ/NAT question types with appropriate scoring (including negative marking).
• Strengths & weaknesses analytics — per-topic accuracy and performance breakdowns to guide what to study next.
• Secure auth — register with email or mobile number, OTP verification (email via Resend, SMS console-stub), JWT access/refresh tokens, forgot/reset password via emailed link.

Tech stack

• Backend: FastAPI, SQLAlchemy 2.0, PostgreSQL, JWT (python-jose), bcrypt
• Frontend: React 18, TypeScript, Vite, Tailwind CSS, React Router, TanStack Query, Zustand, Recharts
• Content pipeline: Python (pypdfium2 for PDF text extraction, custom multi-format parsers per exam)

Project structure

GRIT/
├── backend/            FastAPI app (models, routes, schemas, auth, scoring)
├── frontend/
│   ├── student_app/    React student-facing web app
│   └── admin_panel/    React admin panel (question management)
├── database/schema/    SQL schema + migrations (run in order, see docs/)
├── imports/            PDF-to-database content pipeline (parsers + importers)
├── content/            Parsed question corpus (markdown, one file per question) + metadata CSVs
└── docs/               Setup guide

Getting started

1. Set up PostgreSQL and run the schema files in database/schema/ (see docs/SETUP_GUIDE.md for a detailed pgAdmin walkthrough).
2. Copy backend/.env.example → backend/.env and imports/.env.example → imports/.env, filling in your local database credentials.
3. cd backend && pip install -r requirements.txt && uvicorn main:app --reload
4. cd frontend/student_app && npm install && npm run dev
5. Import question content: cd imports && python import_questions.py --all

Status

Actively in development. GATE corpus covers 2020–2026 across 14 branches; CAT corpus covers 2021–2024 (2017–2020 not yet parsed). AI personalization (Phase 1) is live; richer GenAI-based coaching is planned.