mina khoshbazmf

Built machine learning risk models and reproducible Python/R data pipelines for national-scale HPAI outbreak risk assessment across Canada, integrating multi-source datasets and engineering features. Validated reliability using temporal, province, grid-cell, and spatial-block holdouts and developed a public web-based decision-support dashboard for near-real-time outbreak monitoring and interactive

Developed a .NET Core certificate authority system supporting 100k+ requests per month with 99.9% uptime, including performance validation for 10,000+ concurrent users. Contributed to agile, production-oriented engineering workflows focused on iterative delivery and reliability, and led technical knowledge-sharing on design, implementation practices, and system reliability.

Conducted industry research on 5G, edge computing, and vehicular networks, developing optimization and deep reinforcement learning methods for task offloading in vehicular edge computing. Evaluated models via simulation-based experiments targeting reduced processing delay and energy-consumption objectives with guidance from an industry mentor.

Developed web modules, workflows, forms, and dashboards for a digital waste management system, including backend development and workflow automation supporting 3,000+ monthly users. Ensured reliability under peak traffic through performance and stress testing and accelerated workflow cycles by 40% via process automation and digitized approvals.

PhD research building ML-based risk models and reproducible data pipelines for multi-source outbreak risk assessment, including validation, interpretable outputs, and decision-support systems.