Summary
Kyle Dorman is a Machine Learning Engineer with 12 years of experience building production computer vision systems across retail, robotics, and e-commerce. As the first engineering hire at a YC-backed startup, he helped scale the company to unicorn status and repeatedly takes products from zero-to-one—prototyping models and leading teams to mature, robust deployments. He has deep hands-on expertise in pose estimation, multi-view 6D pose and reprojection refinement, real-time multi-camera tracking, and retail shelf detection, often combining open-source research (GroundingDINO, Segment Anything) with custom tooling for auto-labeling and deployment. Kyle excels at closing the sim-to-real gap through synthetic-data workflows and clever augmentation, and has driven measurable production gains (e.g., 2x inference speed, 50% fewer identity swaps, sub-millimeter pose accuracy). Based in San Francisco, he is motivated by efficient, creative solutions that deliver real-world impact and enjoys tackling the messy engineering problems that take prototypes into reliable systems.
12 years of coding experience
9 years of employment as a software developer
Biological and Environmental Engineering, Biological and Environmental Engineering at Cornell University