Hello,
I am Alexander.
Education
University of Toronto (BASc in Engineering Science)
- Machine Learning and Aritificial Intelligence Major
- Engineering Business Minor
Upper Canada College (International Baccalaureate)
- Top 5% of Class (2015-2019)
- General Proficiency Award (2015-2019)
Previous Experiences
Here are some companies I have worked for in the past!



- Built AI data extraction pipelines with Django and Python to process documents like passports and other formatted and unformatted files, achieving over 99% accuracy in data extraction and enabling automated candidate screening.
- Developed responsive front-end components for data extractors using TypeScript and React, improving data population speed by over 4x for users.
- Created an internal cross-validation tool for extraction, using SQL databases and integrating it with Django’s ORM; this accelerated the iterative development process by 5x and allowed for precise performance measurement across different extraction methods.
- Used Java in an agile environment with the Matching Engine Trading Team to maintain and build new features into low-level ultra-low latency protocols, on RESTful API and message-oriented systems.
- Developed and deployed into production end-of-day protocol that aggregated all activity for the day for Dubai Gold & Commodities Exchange which allowed for 90% reduction in information retrieval time.
- Developing unit tests with JUnit and Mockito, ensuring secure, robust, and ultra low latency programming, and using Docker, and the Nasdaq Validator to test and validate.
Researcher (2021-2022)




*Patient Information is not shown due to confidentiality reasons*
- Publication: “Predicting Tumor Progression in Patients with Cervical Cancer Using Computer Tomography Radiomic Features.” MDPI Radiation 2024, 4, 355-368. https://doi.org/10.3390/radiation4040027
- Deep learning approaches to classification of breast QUS spectral parametric maps using transfer learning with Python (Tensorflow, and PyTorch).
- Increased efficiency by over 300% in optimal hyperparameter searching, by implementing Bayesian Optimization and Hyperband to narrow down the best performing models for each parametric map, resulting in an average of 10% increase of balanced accuracy for the segmentation of breast ultrasound images, achieving over 85% accuracy on real patient data.
- Radiomic analysis of CT and MRI using machine learning to predict treatment response and clinical outcomes in cervical cancer patients using deep learning models.
Skills Used
- TensorFlow: Used for deep learning implementations.
- PyTorch: Another deep learning library used for transfer learning and classification tasks.
- pyradiomics: Used for the extraction of radiomics features from CT images.
- Python: Used for implementing machine learning models, deep learning frameworks (TensorFlow, PyTorch), feature extraction (pyradiomics), and data preprocessing techniques such as SMOTE and Isolation Forest, and hyperparameter optimization.
- Command Line/Bash Scripting: Used for automating processes like data conversion and handling medical imaging files.
- DICOM (Digital Imaging and Communications in Medicine): Used for handling and storing medical imaging data.
- NIfTI (Neuroimaging Informatics Technology Initiative): Used for image file conversion and analysis.
- CT Imaging: Used as the primary imaging modality for radiomics feature extraction.
Image Registration: Used for aligning imaging data and tumor segmentation.
Production Advisor (Summer 2021)
Enbridge
Working with the technical training team. Primarily responsible for assisting with the development of technical training content.

*Examples are not shown due to confidentiality reasons*
- Data Management and Integration: Successfully merged Enbridge and Union Gas servers into a unified system while resolving over 200 instances of discrepancies in training tools and presentation content, ensuring smooth operations and consistency across platforms.
- Content Development and Production: Led end-to-end creation of training materials, including developing storyboards from training manuals, on-site filming, and editing. Produced a fully developed training video that was approved and implemented as part of the organization’s training program.
- Innovative VR Content Creation: Conducted research and experimentation with 360 VR technology to pioneer immersive and interactive training solutions. Delivered an interactive proof of concept to showcase the potential of this cutting-edge approach.
- Quality Assurance and Research: Evaluated new training content for usability, user-friendliness, and bugs while performing needs assessments and effectiveness evaluations to enhance the overall quality and impact of training materials.