Muhammad Humam

Full-Stack Developer, AI/ML Engineer
& Game Dev Enthusiast

Crafting innovative solutions at the intersection of software, machine learning, cybersecurity, and game development.

I started coding in 2018 by making my first voice assistant using Python tutorials on YouTube. This got me interested in game development, and in 2020 I started learning Unity. Since then, I've worked with many clients around the world like Friscon, YNotAI, and Research Inter at Institue of Space Technology.

My main focus is cybersecurity and machine learning. I use advanced technologies to build secure AI systems. I've done projects like creating encryption algorithms, improving web security, and developing AI models for general purpose, healthcare & finance.

While cybersecurity and machine learning are my main areas, I still enjoy game development as a hobby. In my free time, I like exploring new game design ideas, watching anime, going to the gym, and spending time with family and friends. I'm always learning and looking for new challenges in technology. My different interests have helped me become a creative problem-solver.

Experience

Sept

Oct 2023

FullStack developer - Contract
ATH RealEstate - DE

Built property listing website using NextJS in one week under tight deadline.

NextJS

TailwindCSS

Javascript

July

Sept 2023

Researcher - Intern
Institute of Space Technology - PK

Led team doing literature review on entrepreneurial intention. Conducted survey of 200 students.

Research

Data collection

Survey

June 2023

Jan 2024

Mobile App Developer - Contract
Friscon - US

Built e-commerce app with React Native, Firebase, and Stripe payments. Used Agile methods.

React Native

Stripe

Firebase

Javascript

May

Dec 2023

UI/UX Designer - Contract
BuildMate - PK

Created web design using Figma while collaborating with a team of 10.

Figma

UI/UX Design

View Full CV

Selected Projects

Image Description Gen AI

AIML Engineer
Image Description Gen AI

Developed an AI model for image description using a dataset of 20 million images. Designed and optimized model architectures including ResNet, LSTM, and Transformer for detailed caption generation. Achieved a maximum ROUGE score of 40%.

Computer Vision

NLP

AIML

Model Architecture Designing

Python

PyTorch

OCR for Handwritten Text

AIML Engineer
OCR for Handwritten Text

Developed a custom OCR system with an accuracy of 92% on a dataset of 800,000 images. Performed comprehensive data augmentation and sampling to enhance model robustness and accuracy.

Computer Vision

AIML

Python

PyTorch

Crypto / Stocks Prediction AI

AIML Engineer
Crypto / Stocks Prediction AI

Developed and optimized a crypto coin and stock price predictor, with accuracy of 88%. Created a LSTM model to predict value based on previous data and time.

AIML

Python

TensorFlow

E-Bazaar

FullStack Developer
E-Bazaar

Led a team of 3 in developing a full-stack E-Commerce platform with integrated auction, wholesale, and marketplace functionalities using MongoDB, Go, and NextJS. Achieved less than 600ms page load time.

FullStack

Go Lang

NextJS

MongoDB

CashFlow

FullStack & Mobile App Developer
CashFlow

Led a team of 2 to build a secure web and mobile finance app using NextJS and React Native. Implemented a secured Python back end and utilized a JSON database as an API, ensuring sensitive data integrity and protection.

FullStack

Mobile App

React Native

NextJS

Python

Inspired by Brittany Chiang, coded in NVIM. Built with Next JS and Tailwind CSS deployed with Vercel. All text is set in the Inter typeface & color theme is Catppuccin.