Summary
Brian Holestine is an AI Engineer specializing in computer vision and machine learning with over a decade designing perception systems and eight years of professional experience deploying solutions across industry and research. He has built end-to-end vision pipelines—from camera selection and data collection to YOLO-based detection, SORT tracking, and event detection for live AR sports experiences—and currently applies ML to interpret electron-beam imagery at Hitachi High-Tech America. At Intel he developed a compact, high-speed self-supervised interest-point model for visual odometry and integrated pose estimation into volumetric video, demonstrating a knack for squeezing state-of-the-art performance into resource-constrained environments. As an instructor and mentor he distills complex topics (ML, ViT, GANs, time series) into practical notebooks and demos, including YOLO fine-tuning and Hungarian-algorithm tracking. Based in Hillsboro, OR, he combines deep academic foundations (MS ECE) with hands-on production chops across vision, robotics, and manufacturing analytics. An often-overlooked strength is his early full-stack tooling work—converting legacy codebases and building visualization and automation tools—that helps bridge research prototypes to maintainable products.
8 years of coding experience
21 years of employment as a software developer
MS Electrical & Computer Engineering, MS Electrical & Computer Engineering at Portland State University
Natural Language Processing Nanodegree Deep Learning, Natural Language Processing Nanodegree Deep Learning at Udacity