Yiming Chen

Machine Learning Engineer | Computer Vision | JHU MSE

Washington DC-Baltimore Area

About

I am a Machine Learning engineer-in-training and M.S.E. candidate in Computer Science at Johns Hopkins University, with a strong focus on computer vision and intelligent systems. My work explores the intersection of advanced learning architectures and practical system design. I build end-to-end intelligent systems — from model architecture and data strategy to scalable software and user-facing applications. I am particularly interested in visual understanding and video modeling, where structured representation learning must operate under real-world uncertainty. My experience spans deep learning–based detection and segmentation, interpretable clinical ML systems, 3D geometric reconstruction, and real-time collaborative application development. I primarily work with Python, PyTorch, and C++, and have experience integrating ML pipelines with modern web frameworks and cloud-based synchronization tools. I am currently seeking Machine Learning Engineer or applied AI roles.

Experience

  • The Johns Hopkins University (Part-time · 1 yr 1 mo)
    • Student Researcher
      Jun 2025 - Present · 1 yr 1 mo

      - Designed and maintained an end-to-end transformer-based temporal action localization system for surgical workflow recognition. - Re-architected the proposal generation pipeline by replacing heuristic sliding-window approaches with dense learned temporal predictions, significantly improving coverage and recall. - Led model optimization and architectural refinement efforts to improve localization robustness across multiple temporal IoU thresholds. - Addressed long-tail class imbalance using logit-adjustment techniques, improving minority-class stability while preserving head-class performance.

    • Course Assistant
      Feb 2026 - Present · 5 mos

      - Led office hours and provided debugging guidance for PyTorch/OpenCV-based training and evaluation pipelines across course assignments. - Supported students in understanding classical computer vision foundations (e.g., camera models, projective geometry) and modern deep learning architectures by translating theoretical concepts into practical implementation strategies.

  • Research Assistant at Jinan University
    Jul 2022 - Jul 2023 · 1 yr 1 mo

    - Improved YOLOv7 detection robustness under adverse-weather degradation through adaptive feature filtering and enhanced preprocessing strategies. - Designed physically-constrained style-transfer data augmentation to simulate distribution shifts and improve model generalization.

  • Software Engineer Intern at Shenzhen Fang Yuan Bao Information Technology Services Co.,
    Jul 2022 - Aug 2022 · 2 mos

    - Contributed to the development of a network acceleration system for online games, focusing on routing optimization and traffic configuration workflows. - Built automated utilities to extract runtime metadata (game progression name, accelerator node IP/port, routing tables) and generate structured XML configuration files. - Conducted functional testing and validation of accelerator performance across multiple network conditions.

  • Machine Learning Engineer at Jinan University
    Sep 2021 - Mar 2022 · 7 mos

    - Designed and deployed a decision-tree–based clinical pre-diagnosis system for urolithiasis risk screening in a hospital setting. - Structured clinical data pipelines and developed a Vue.js-based interface for patient intake and risk assessment. - Led model design and feature engineering to improve interpretability and practical usability in real-world clinical workflows.