Joel Carriere

Frameworks & Libraries: sci-kit learn, Pandas, NumPy, Matplotlib, PyTorch, OpenCV

Machine Learning Models: Deep Neural Networks, Support Vector Machines, Decision Trees, KNN, Regression

Machine Learning Projects: Biomechanical angle determination, pose estimation, face detection, face classification

Master of Applied Science, Specialization in Artificial Intelligence
University of Guelph, Guelph, Ontario | GPA 3.8/4.0
Sept 2021 - Dec 2023

Bachelor of Engineering, Biomedical (Co-op);
Minor: Business Administration
University of Guelph, Guelph, Ontario
Sept 2014 - April 2019

Golf AI
June 2024 - Present

Lead the development of a mobile and web application tailored to the golf business.
Developed algorithms and provided top-notch algorithms to provide golfers with low-cost, highly accurate data from their phone/computer.
Contributed to the development of business plans, brand strategy, brand identity, and overall targets to ensure targets were met.
Led the development of a robust data processing pipeline to provide golfers, coaches, and businesses with a snapshot of metrics to improve success.
Designed experiments to collect data to better improve overall success of the models being deployed on the mobile and web apps.

University of Guelph, Guelph, Ontario
Sept 2021 - Present

Engineered an innovative proprietary solution for the determination of human wrist ergonomics. Achieved accuracy and precision comparable to industry gold standards; a significant advancement in ergonomic research.
Conducted extensive testing on various model types to optimize wrist posture angle prediction from 2D data, achieving a competitive Mean Absolute Error (MAE) of 7 degrees. Further improved accuracy with the integration of 3D datasets, achieving an impressive MAE of 5 degrees, aligning with top-tier markerless motion capture systems.
Employed cost-effective off-the-shelf equipment in controlled experiments, and open-source software to capture and generate an extensive dataset of human wrist motion data. This was crucial for successful machine learning model training, enabling predictive analysis of ergonomic wrist postures through computer vision techniques.
Led the development of a robust pipeline combining YOLOv5 and ResNet-34 Convolutional Neural Networks (CNN) to process extensive open-source datasets, extract facial features from images, and perform gender and race classification. Achieving 99% confidence in gender classification and an impressive 95% confidence in race classification.
Designed experiments to evaluate machine learning models on a multi-modal human action dataset achieving an F1 score of 0.90 in classifying 21 upper body motions related to biomechanical movement.

Sonova Group, Kitchener
Nov 2020 - August 2021, Feb 2023 - Oct 2023

Played a pivotal role in orchestrating the successful launch of hearing devices and overseeing post-launch engineering activities, ensuring seamless transitions and optimal product performance.
Supervised and conducted equipment maintenance activities in the research and development area, maintaining a conducive operational environment.
Worked closely within a collaborative team, ensuring the alignment of project timelines and deliverables to guarantee the timely launch of new hearing devices and achieve organizational goals.

IBM Databases and SQL for Data Science with Python
Completed November 2023

Responsive Web Design - freecodecamp
Completed September 2022

Lean Six Sigma Green Belt - Abacus Team
Completed August 2020