Hi, I'm Usama 👋
Aspiring Python developer with a focus on robotics, AI, and data science. I enjoy designing and coding automated solutions that make a real impact.
UM

About

Honors BCS student concentrating in Big Data at Wilfrid Laurier University. Since 2022 I have balanced research, internships, and teaching while building ML and full stack projects. Recent work includes automating workflows with Power Platform and Azure, building data pipelines in Python, and shipping responsive e-commerce experiences. I also conduct research in adversarial ML and robotics using Jetson Nano and Yahboom platforms.

Work Experience

G

Government of Ontario

September 2024 - Present
Software Developer Intern
Automated journal entries with Power Automate Desktop and Azure Logic Apps reducing processing time by 40 percent. Built Power Apps with Dataverse and SQL improving data management efficiency by 35 percent.
S

Shoponomik

May 2024 - Present
Software Engineer Intern
Developed a logger for an e-commerce platform and boosted logging efficiency by 50 percent using Selenium test automation. Cut data retrieval time by 40 percent by building optimized Python data pipelines.
S

Sports.Excitement

September 2023 - December 2023
Full Stack Developer Intern
Shipped a responsive e-commerce site using React and Tailwind CSS improving accessibility by 40 percent. Increased checkout speed by 30 percent using PHP Laravel while collaborating with a 20 person team.
U

University of Waterloo

May 2023 - August 2023
Machine Learning Research Assistant
Reviewed 10 plus papers on adversarial attacks and defenses with a focus on patch based strategies. Built a CNN using five adversarial attacks and patches to improve classification accuracy. Explored RTX ray tracing cores and outlined three applications in real time rendering and AI.
W

Wilfrid Laurier University

September 2022 - Present
Undergraduate Researcher and Instructional Assistant
Led a $10,000 research project using Jetson Nano and Yahboom robots to develop AI that enhances robot functionality and task performance. Supported five professors with lesson delivery and grading. Provided one on one help to 300 plus students across Python, Java, C, Digital Circuits, VBA, Algorithms, Object Oriented Programming, and Data Structures.

Skills

Python
Java
JavaScript
TypeScript
C
R
SQL
VBA
PHP
PLC
NumPy
Pandas
Matplotlib
Seaborn
OpenCV
PyTorch
TensorFlow
Keras
Scikit-learn
TensorRT
Gradio
Jupyter
Google Colab
React
React Native
Node.js
Flask
Tailwind CSS
Bootstrap
Git
Linux
Raspberry Pi
Arduino
ROS
Azure
Power Automate
Power Apps
Dataverse
SQL Server
Selenium
Google Maps Platform
Flutter
Dart
MS Office
My Projects

Check out my latest work

I've worked on a variety of projects, from simple websites to complex web applications. Here are a few of my favorites.

Stock Market Trading Bot

Built a CNN for stock prediction using ResNet, InceptionV3, and DenseNet121. Improved accuracy by 70 percent and sped up training by 20 percent with min max normalization and better handling of missing data. Added a Gradio UI for real time predictions and visualizations which increased engagement by 30 percent.

Python
TensorFlow
Keras
Pandas
Seaborn
Gradio
PIL

Object Recognition Model

Implemented YOLOv4 based detection with 95 percent accuracy. Reduced training time by 20 percent through data augmentation and efficient GPU use. Integrated TensorRT for real time inference with a 30 percent speedup.

Python
OpenCV
Scikit-learn
Matplotlib
TensorRT
SIFT
Stitcher

Book Recommendation Model

Created a hybrid recommender combining collaborative and content based filtering. Increased recommendation diversity by 30 percent through genre analysis and cut training time by 25 percent with efficient feature extraction and normalization.

Python
NumPy
Pandas
Scikit-learn
Difflib
Hackathons

I like building things

During my time in university, I attended 4+ hackathons. People from around the country would come together and build incredible things in 2-3 days. It was eye-opening to see the endless possibilities brought to life by a group of motivated and passionate individuals.

  • J

    JUNR

    Waterloo, ON

    Networking tool that shares social links via QR codes and Jackal wallet addresses. Pivoted from React Native to a web app due to integration constraints and shipped a working proof of concept.
  • I

    In a Flash!

    London, ON

    Flashcard builder and quiz tool using Cohere text generation. Implemented UI and assisted with Node.js backend and first time API integration.
  • N

    Nzyme

    Toronto, ON

    Summarizes textbook chapters with NLP and supports active recall. React frontend connected to a Flask Python backend using Cohere for text generation and entity extraction.
  • C

    Car Less

    Waterloo, ON

    Pokemon GO inspired transit gamification app to encourage public transport use. Built with Flutter and Google Maps Platform APIs.
Publications

Research Contributions

Some of my published research in AI, ML, and Robotics — presented at international conferences and journals.

Real-Time Plant Disease Detection Using YOLOv8 on a Low-Resource Device

Usama Mohiuddin, Emad A. Mohammed 38th Canadian Conference on Artificial Intelligence (2025)

Developed an AI-powered robotic system for real-time plant disease detection using YOLOv8, achieving 90% accuracy on edge hardware (Jetson Orin NX). Integrated robotic spraying and cloud monitoring for greenhouse sustainability.

YOLOv8
Edge AI
Robotics
Computer Vision

Securing 3D Deep Learning Models: Simple and Effective Defense Against Adversarial Attacks

Rosina F. Kharal, Saif Al-Din Ali, Usama Mohiuddin, Samir El Sayed 2024 IEEE International Conference on Machine Learning and Applications (ICMLA)

Examined vulnerabilities of 3D deep learning models under adversarial attacks. Proposed a simplified defense that reduced retraining by ~90% while restoring or exceeding baseline accuracy.

3D Deep Learning
Adversarial ML
Security
Defense
Contact

Get in Touch

Want to chat? Just shoot me a dm with a direct question on LinkedIn and I'll respond whenever I can. I will ignore all soliciting.