Hello, I'm
Zhongqian (Henry) Duan

  • 🔭 MS in Computer Science at University of Michigan
  • ☁️ Software Engineer at @AWS
  • 🕹 Ex Software Engineer at @Rec Room
  • 💻 Ex Machine Learning Engineer intern at @Apple & @NIO
  • 📚 Dual degree holder from University of Michigan (BS in Computer Science @umich) and Shanghai Jiao Tong University (BS in Electrical and Computer Engineering @situ)
  • 🍁 Fun fact: I love 📸 🏓 🍜 🎮

My areas of interest include software (game) development, machine learning and HPC. I aim to deepen my understanding in various areas of technology and product development.

Work Experience

Software Engineer at Amazon Web Service

  • Leveraged AWS services such as Lambda for serverless execution, S3 for log storage, and IAM for secure access management.
  • Designed and implemented a flow operations bot to automate log retrieval from CloudWatch, streamlining troubleshooting and monitoring processes.
  • Expanded bot functionality by integrating support for multiple ticket types, significantly reducing on-call workload and improving operational efficiency.

Software Engineer at Rec Room

  • Worked on the UGC > Logic team for the virtual reality gaming startup.
  • Create an undo system for the maker pen in wiring chips that guarantees the accurate restoration of circuit connections upon undoing wire-related actions. (The circuit is a visual programming tool for UGC content [link]).

Software Engineer Intern at Rec Room

  • Developed over 20 chips using C# and Git version control for Rec Room's CircuitV2 System, including Synced Delay Chip and Player Gaze Chip.
  • Created a tool prefab named "Motion Trail," which utilized Protobuf for Serialization and Photon Unity Networking (PUN) for Synchronization. The purpose of this tool was to enhance the visual appeal of moving objects.
  • Deploy builds on Steam using Jenkins, and collaborated with a team of developers and designers to playtest new features in both screen and VR mode.

Machine Learning Engineer Intern at Apple Inc.

  • Developed a distributed training process using gRPC, featuring a remote validator and data server, which increased the data processing speed by around 10%.
  • Constructed a robust pipeline to create real-time physical simulation datasets, and implemented a dedicated training monitor for self-learning and reinforcement.
  • The generated base model that can be efficiently fine-tuned for specific task, reducing the need for extensive data and enhancing performance.

Machine Learning Engineer Intern at NIO

  • Participated in the NIO black-box project that tests the entire autonomous driving system with on-screen video.
  • Participated in the design and optimization of a 3D Object Tracking Network for autopilot, improved the precision by 4%.
  • Utilized pre-trained vision models, such as MaskRCNN, with different backbones to detect vehicles and lane lines, and tested for autonomous driving systems on over 2000 on-screen videos.
  • Proposed a light-weight CNN with different backbones in PyTorch to remove the lens distortion effect.

Graduate Student Instructor

  • EECS 587 Parallel Computing Course GSI
  • Led weekly discussion sessions and office hours for EECS 587 Parallel Computing. (CUDA, OpenMP, MPI)

Selected Projects

Game Development

Research

Text-Image Pair Generation via Pre-trained Vision-Language Models

[github] [website]

Web Design

A simple online store.

[Link]

Education

Interest

Get In Touch

You are welcome to drop me a message :)

  • Address

    757 Thomas St
    Seattle, WA 98109
    United States
  • Phone

    (+1)734-546-0202
  • Email

    duanzq@umich.edu