The Athlete
Intensity & Discipline
The Engineer
Precision & Logic
Pitch meets Processor.
My journey bridges two worlds: the intensity of competitive athletics and the innovation of software engineering.
On the pitch, I compete. Off it, I build AI systems. This isn't just what I do, it's who I am.
Mehdi
Essoussi
82
Senior Appearances
10th
Overall Pick (2025)
Varsity Captain
Named team captain of U of T Varsity Blues (2024), leading the squad to OUA East accolades.
All-Canadian
U SPORTS Second Team All-Canadian (2025) & 2x OUA East First Team All-Star selection.
CPL Draft Pick
Selected 10th Overall by Vancouver FC in the 2025 CPL-U Sports Draft.
The Journey
A timeline of my soccer career from youth development to professional leagues.
2025
2025
Vancouver FC (CPL)
2025 — PRESENTCanadian Premier League • U-Sports Contract
Signed in March 2025 after being selected 10th overall. Made 15 appearances in the 2025 season before returning to university, with rights retained for 2026.
2023
2023
University of Toronto
2023 — PRESENTOUA / U SPORTS
Impact midfielder and Team Captain (2024). Earned consecutive OUA East First Team All-Star honors (2024, 2025) and U SPORTS Second Team All-Canadian (2025).
2023
2023
Alliance United FC
2023 — 2024League1 Ontario
Registered 30 appearances and scored 1 goal across two seasons, maintaining high-performance levels in the semi-pro circuit.
2017
2017
Toronto FC II & Academy
2017 — 2022USL League One / MLS Next Pro
- Signed first pro contract in 2018. Made 24 appearances for TFC II.
- Signed short-term loan with MLS First Team (May 2022).
- League1 Ontario Second Team All-Star (2018) with TFC III.
Youth
Youth
Youth Foundation
Born in Tunis. Developed at Nepean Hotspurs, Ottawa South United, and Ottawa Fury Academy before joining TFC Academy. Attended Canada U-15 National Camp (2016).
Career Statistics
| Club | League | Apps | Goals |
|---|---|---|---|
| Vancouver FC | Canadian Premier League | 15 | 0 |
| Alliance United | League1 Ontario | 30 | 1 |
| Toronto FC II | USL / MLS Next Pro | 24 | 0 |
| Toronto FC III | League1 Ontario | 13 | 1 |
| CAREER TOTAL | 82 | 2 | |
Career in Pictures
Highlight Reels
Watch my career highlights and memorable moments on the pitch.
Mehdi Essoussi
Computer Engineer
class Mehdi(Athlete, Engineer):
def __init__(self):
self.passion = "building intelligent systems"
self.superpower = "turning ideas into reality"
self.fuel = ['coffee', 'curiosity', 'competition']
# Implementation plan
def approach_problem(self, challenge):
return self.analyze() + self.execute() + self.iterate()
➜ always learning, always building...
Technical Skills
Python
C/C++
Java
SQL
JavaScript
TypeScript
React
Next.js
Node.js
Flask
FastAPI
PyTorch
TensorFlow
NumPy
Supervision
Docker
AWS
Git
Education
University of Toronto
Bachelor of Applied Science (B.A.Sc), Electrical and Computer Engineering
Toronto, ON
Graduated June 2025
Professional Experience
AI / Computer-Vision Engineer (Project Lead)
June 2025 – PresentNorth of Zero • Toronto, ON
- Architected a modular video-analytics pipeline converting raw footage to pro-stats in ≤10 min.
- Built end-to-end detection → tracking stack (fine-tuned YOLO + DeepSORT) in Docker, slashing latency to <40 ms.
- Led a 3-member cross-functional team, defining sprint backlogs and coordinating PhD-level reviews.
IT Support Specialist
June 2024 – May 202530 Forensic Engineering • Toronto, ON
- Built an AI-powered report automation tool, reducing drafting time by 20% for forensic engineers.
- Optimized internal database workflows and implemented process improvements via automation.
- Developed technical documentation and provided support for VOIP/Fob systems.
Featured Projects
Soccer Analytics System
AI-driven detection and tracking framework processing full matches with real-time analytics.
Soccer Analytics System
Containerised detection-tracking framework (YOLOv8, ByteTrack) in Docker processing 90-min matches on single GPU. Integrated Label Studio for custom datasets.
Overview
Built an AI-driven soccer analytics system with detection and tracking capabilities. Containerised the entire framework using Docker for seamless deployment and GPU acceleration.
Integrated computer vision models including YOLOv8 for player detection and ByteTrack for multi-object tracking across 90-minute matches. Set up Label Studio annotation platform for creating and managing custom training datasets.
Key Features
Technologies & Tools
AI Report Writing App
Automated forensic engineering reports with AI agents, improving efficiency by 20%.
AI Report Writing App
AI-driven tool automating forensic engineering reports with structured document generation and intelligent agents.
Overview
Built an intelligent document generation system specifically designed for forensic engineering reports, utilizing AI agents to streamline the reporting workflow.
Developed an AI-driven tool to automate forensic engineering report drafting, reducing drafting time by 20%. Implemented AI agents for structured document generation, improving workflow efficiency.
Key Features
Technologies & Tools
Mechanical Parts Classifier
Deep learning model classifying mechanical components using CNNs trained from scratch.
Mechanical Parts Classifier
Custom CNN architecture trained from scratch to classify mechanical components with high accuracy across multiple part categories.
Overview
Developed a deep learning model for classifying mechanical parts using convolutional neural networks. Built and trained custom CNN architectures from scratch using both TensorFlow and PyTorch frameworks.
Implemented data augmentation techniques and transfer learning strategies to improve model performance. The classifier achieves high accuracy in distinguishing between various mechanical components and parts categories.
Key Features
Technologies & Tools
UofT Second Hand Hub
Peer-to-peer marketplace for students with scalable cloud architecture and optimizations.
UofT Second Hand Hub
Full-stack peer-to-peer marketplace platform built for university students, featuring scalable cloud infrastructure and optimized performance.
Overview
Built a comprehensive peer-to-peer marketplace web application designed specifically for university students to buy and sell second-hand items. Developed with a focus on scalability and user experience.
Implemented cloud-based architecture using AWS services for reliable hosting and storage. Created a responsive React frontend with SQL database backend, optimizing query performance and implementing efficient data management strategies.
Key Features
Technologies & Tools
E&M Field Visualizer
Interactive tool visualizing electromagnetic concepts with real-time animations.
E&M Field Visualizer
Interactive web-based tool for visualizing electromagnetic field concepts with real-time animations and physics simulations.
Overview
Created an educational visualization tool to help students understand electromagnetic field concepts through interactive simulations. Built using React and PixiJS for high-performance graphics rendering.
Implemented real-time physics calculations and animations to demonstrate electric and magnetic field interactions. The tool provides an intuitive interface for exploring complex electromagnetic phenomena with visual feedback.
Key Features
Technologies & Tools
Knock-Activated Door Lock
IoT-based smart lock using secret knock sequences for authentication.
Knock-Activated Door Lock
IoT-based smart lock system using pattern recognition to authenticate users through secret knock sequences on STM32 microcontroller.
Overview
Developed an innovative IoT-based door lock system that uses pattern recognition to authenticate users through unique knock sequences. Built on STM32 microcontroller platform with C++ programming.
Implemented signal processing algorithms to detect and analyze knock patterns with high accuracy. The system features low-power operation, real-time pattern matching, and secure authentication through customizable knock sequences.
Key Features
Technologies & Tools
City Map Location Tool
Web-based map tool with Google Maps-like features and custom team collaboration.
City Map Location Tool
Feature-rich GIS application with Google Maps-like functionality, pathfinding algorithms, and collaborative team features for urban navigation.
Overview
Built a comprehensive web-based mapping application using C++ and GTK framework. Implemented advanced pathfinding algorithms including A* for efficient route calculation and navigation across city maps.
Developed Google Maps-like features including interactive map exploration, location search, route planning, and custom markers. Added team collaboration features for shared map views and synchronized navigation experiences.
Key Features
Technologies & Tools
Trinity College Rooftop
Sustainable rooftop design optimizing environmental impact for Trinity College.
Trinity College Rooftop
Comprehensive sustainable rooftop design project for Trinity College, focusing on environmental optimization and green architecture principles.
Overview
Designed a sustainable rooftop solution for Trinity College using AutoCAD and sustainable architecture principles. Focused on minimizing environmental impact while maximizing functional space utilization.
Implemented green design strategies including rainwater harvesting, solar panel integration, and native plant selection. The design balances aesthetic appeal with environmental responsibility and long-term sustainability goals.