About Me
I’m Mohammad Hedayati, an applied ML researcher and data scientist based in Toronto. My work sits at the intersection of advanced algorithms and real‑world impact: from building deep‑learning models that catch subtle anomalies in satellite subsystems to designing prediction engines that guide multi‑million‑dollar fundraising campaigns.
What I do today
I turn messy, high‑volume data into actionable insight. Recently, I engineered a Temporal Convolutional Network (TCN) and an optimization layer for DonorVoice, boosting donation‑probability forecasts and helping charities target outreach while saving on ad spend. Before that, my graduate research produced a hybrid ResNet + Particle‑Filter framework that achieved 99 % accuracy in detecting reaction‑wheel faults—work published in Acta Astronautica and presented at international PHM conferences.
Why it matters
Predictive analytics isn’t just a buzzword; it’s a way to cut costs, reduce risk, and uncover hidden opportunities. I enjoy taking projects from whiteboard to production, validating them with hard numbers, and sharing the results through peer‑reviewed papers, open‑source notebooks, or straightforward dashboards that non‑technical teams can trust.
Beyond the code
I grew up in Tehran and earned my Aerospace B.Sc. at Sharif University before moving to Canada for an M.A.Sc. in Mechanical Engineering. Bilingual in English and Farsi, I’m comfortable working across cultures. When I’m not prototyping models, you’ll likely find me sketching ideas for the next R&D challenge, working my way through a fresh stack of films and books, or getting outside for a game or a run.
Projects
A Novel Framework for Real-Time Condition Monitoring Using Signal Streams:
1D Sliding Window Residual Network (ResNet)
A Novel Particle Filter Algorithm:
Markov Jump-Adjusted Particle Filter (MJAPF)
Generating Realistic Time-Series Datasets Using Wasserstein Generative Adversarial Networks (WGANs) and Dynamic Time Warping (DTW)
An Auxiliary Particle Filter (APF) Implementation
Publications
Hedayati M., Rahimi A. "A hybrid framework for real-time satellite fault diagnosis using Markov jump-adjusted models and 1D sliding window Residual Networks", Acta Astronautica (2025)
Hedayati M., Barzegar A., Rahimi A. "Fault Diagnosis and Prognosis of Satellites and Unmanned Aerial Vehicles: A Review", Applied Sciences (2024)
Hedayati M., Barzegar A., Rahimi A. "Mitigating Data Scarcity for Satellite Reaction Wheel Fault Diagnosis with Wasserstein Generative Adversarial Networks", Proceedings of 2024 IEEE Conference on Prognostics and Health Management (ICPHM 2024)