AVAILABLE FOR INTERNSHIP

Muhammad
Ahmed
Zia.

Data Science & Machine Learning Enthusiast

CS student at COMSATS University Islamabad. I build end-to-end ML systems — from data collection and model training to Flask REST APIs and cloud deployment. Passionate about computer vision, NLP, and intelligent system design.

Contact Me ↗ Download CV ↗

Projects Built

98.26%

Best Model Accuracy

2023

COMSATS CS Student

OPEN TO WORK

University

COMSATS · Lahore

SCROLL keyboard_arrow_down

Featured Work

Projects

View GitHub ↗

Crop Doctor

AI Crop Disease Diagnosis System · 92% Accuracy

AI-powered web system for Pakistani farmers to diagnose crop diseases from a photo or Urdu/English symptom description. EfficientNet (CV) + multilingual BERT (NLP). Covers 5 major Pakistani crops with actionable treatment recommendations.

Python EfficientNet mBERT Flask HTML/CSS/JS

Health Risk AI

Lifestyle-Based Risk Predictor · 88.3% Accuracy

Random Forest classifier on a 5,000-record lifestyle dataset. Benchmarked against Decision Tree (~82%) and Naive Bayes (~78%). Flask REST API with gauge chart, factor breakdown, and personalised health advice. Deployed on Vercel.

Random Forest scikit-learn Flask Docker Vercel

PhishGuard AI

Email Phishing Detection · 98.26% Accuracy

Hybrid ML + heuristic system trained on 18,000+ labeled emails. TF-IDF + Logistic Regression pipeline with live Gmail OAuth scanner — scans 50 emails in under 2 seconds using concurrent threading.

scikit-learn TF-IDF Flask Gmail OAuth Python

E-Commerce App

Flutter + Firebase + Kubernetes on Azure

Full-featured cross-platform mobile app with Seller and Buyer dashboards for product, cart, and order management. Firebase Auth + Firestore real-time sync. Containerised with Docker, deployed on Kubernetes via Azure.

Flutter Dart Firebase Kubernetes Azure

Vibee

Centralized Social Media Aggregator

Cross-platform Flutter app unifying YouTube, Reddit, and News into one personalised feed. Interest-to-content mapping for 16 categories, timed screen sessions, skeleton loading animations, and custom Firestore data model.

Flutter YouTube API Reddit API Firebase NewsAPI

My Capabilities

Technical Arsenal

A curated selection of technologies I use to build high-performance, scalable AI systems and meaningful user experiences.

psychology

Data Science & ML

AI · NLP · Computer Vision

scikit-learn Random Forest EfficientNet (CV) multilingual BERT RAG Pipelines Model Evaluation Cross-Validation NumPy Pandas ChromaDB sentence-transformers

code

Languages

Core · Scripting · Query

Python Dart Java SQL JavaScript

layers

Frameworks & Libraries

Backend · Mobile · UI

Flask FastAPI Flutter Streamlit joblib

database

Databases & Cloud

Storage · Infra · AI APIs

Firebase Firestore Azure Kubernetes Service Docker MongoDB Google Gemini API

build

Tools & Platforms

DevOps · Deployment · Version Control

Git GitHub REST APIs Vercel Render Heroku Docker Kubernetes HuggingFace Spaces Railway

MuhammadAhmed Zia.