Skip to content
AZKA.
← All projects
Editor's Pick2024

CaptainMDCAT

AI-powered MDCAT preparation platform with adaptive testing, personalized study plans, and performance analytics.

Role
Founder & Developer
Duration
6 Months · 2024
Stack
React · Firebase · Python · TensorFlow
CaptainMDCAT overview - adaptive learning home
Adaptive testing interface

Case Study

CaptainMDCAT

Mobile View

03 / 08

Replace with Screenshot

Product pages

Full-page views

Two tall page compositions escaping past the laptop base.

Adaptive learning home - full page
Performance analytics - full page

01 - Chapter

Overview

CaptainMDCAT is an AI-powered preparation platform for the Medical and Dental College Admission Test (MDCAT). It's not just another question bank-it's a complete study companion that adapts to each student's learning style and pace.

The platform has helped thousands of aspiring medical students prepare smarter, not harder, by leveraging data-driven insights and personalized learning paths.

02 - Chapter

Challenges

MDCAT preparation is notoriously stressful. Students often don't know where to focus, waste time on topics they've already mastered, and lack visibility into their actual readiness for the exam.

The existing solutions were either glorified PDF dumps or expensive coaching centers inaccessible to many. The challenge was democratizing quality MDCAT prep while making it genuinely effective.

03 - Chapter

Architecture

CaptainMDCAT uses a React frontend with Firebase for authentication, real-time data, and Cloud Functions for server-side ML inference. TensorFlow powers the adaptive testing engine, while Algolia provides fast question search. Chart.js renders performance analytics dashboards.

04 - Chapter

User Flow

Students sign up, take a diagnostic test, and receive a personalized study plan. Daily practice sessions adapt in real-time based on performance. Mock tests simulate exam conditions, and leaderboards create healthy competition throughout the prep journey.

05 - Chapter

Development Process

Started with user research among MDCAT aspirants, wireframed the adaptive testing flow, iterated on the study planner algorithm, and deployed as a PWA for offline access. Gamification features were added based on user retention data.

06 - Chapter

Results

Since launch, CaptainMDCAT has served over 10,000 students. The average user sees a 23% improvement in mock test scores within the first month of consistent usage. Students have secured admissions in top medical colleges across Pakistan.

07 - Chapter

Technologies

React, Firebase, Python, TensorFlow, Cloud Functions, Algolia, Chart.js, and PWA technologies power the full stack-from adaptive ML inference to real-time leaderboards.

08 - Chapter

Reflection

Building CaptainMDCAT taught me that AI in education works best when it's invisible-students shouldn't feel like they're using ML, they should feel like the platform understands them. I'd invest earlier in A/B testing the study planner and build more robust offline capabilities.

Gallery

Device frames - click for fullscreen.

Technologies

Tools and platforms behind the build.

React
Firebase
Python
TensorFlow
Cloud Functions
Algolia
Chart.js
PWA