Summary
Duke Lin is a Machine Learning Engineer with nine years of experience applying deep learning, computer vision, and reinforcement learning to production problems, currently building ML systems at Zoox after roles at Meta and Johnson & Johnson. He brings hands-on expertise in PyTorch (including DDP), Transformers, SLAM, and model interpretability, and pairs that with practical MLOps skills using Docker, MLflow, DVC, and cloud infrastructure. Duke’s background in intelligent systems and robotics (MS, UC San Diego) informs a systems-minded approach to feature extraction and maintenance-detection pipelines that balance research rigor with stakeholder needs. He’s comfortable owning the full model lifecycle from data analysis and metric design to deployment, and has a track record mentoring teammates while pushing on cutting-edge architectures. A subtle differentiator is his cross-domain research experience—from academic robotics labs and an international visiting stint to large-scale industry ML—enabling fast adaptation across both research and production contexts.
9 years of coding experience
5 years of employment as a software developer
University of California, San Diego