Amirreza Yasami is a Ph.D. candidate in Mechanical Engineering at the University of Alberta. He holds a Master’s degree in Mechanical Engineering from the University of Tehran. His work focuses on trustworthy machine learning for real-world engineering systems, combining scientific modeling with modern AI to produce deployable and interpretable solutions.

Through Amii, Amirreza works with Dune Engineering on applied AI systems for document intelligence and perception. His work includes Optical Character Recognition (OCR) and Vision-Language Models (VLMs) for interpreting structured engineering drawings and technical documents, supporting floor plan analysis and building energy assessments. He works with both open-source and proprietary multimodal models while considering accuracy, privacy, and deployment cost. His broader goal is to bridge scientific machine learning and operational AI by translating advanced models into reliable, secure, and production-ready engineering tools.

In his academic research, he develops multimodal and physics-informed learning frameworks for predicting vehicle fuel consumption and emissions under highly dynamic operating conditions. He designs advanced temporal architectures integrating sequence modeling, representation learning, and explainability to build virtual emission sensors and next-generation drive-cycle modeling tools.