Nguyen Dzung
Sr. Machine Learning Engineer at Axon
Hà Nội, Vietnam
Summary
🤩
Rockstar🎓
Top SchoolNguyen Dzung is a Senior Machine Learning Engineer with nine years of hands-on experience building real-time, privacy-preserving computer vision systems for autonomous vehicles and enterprise applications. At Axon he architected distributed training and GDPR-compliant data collection and local experimentation platforms that shortened hyperparameter tuning from weeks to days and enabled safe ALPR expansion across multiple countries. His background spans research-grade 3D LiDAR perception and lightweight embedded solutions—from high‑fps 3D detectors and LiDAR cluster classifiers to Jetson-based parking detectors—reflecting both deep model expertise and production engineering. An award-winning KIST master's graduate who ranked in the top 2% of his undergraduate class, he is an active open-source maintainer of PyTorch 3D-detection projects where he refactors code, optimizes training pipelines, and integrates datasets for reproducible research. Notably, he blends algorithmic rigor with infrastructure-as-code and cloud-native deployments, making him effective at moving models from lab prototypes to GDPR-compliant production.
10 years of coding experience
8 years of employment as a software developer
High school, Mathematics, High school, Mathematics at HaTinh high school for gifted studentsBachelor's degree, Electronics and Telecommunications, Distinction, Bachelor's degree, Electronics and Telecommunications, Distinction at Hanoi University of Science and TechnologyNanodegree Program, Sensor Fusion, Nanodegree Program, Sensor Fusion at UdacityMaster's degree, (KIST School) HCI & Robotics, 4.27/4.50, Master's degree, (KIST School) HCI & Robotics, 4.27/4.50 at (UST) University of Science and Technology, Korea
Vietnamese, English, Korean
Stackoverflow
11reputation
6kreached
1answer
0questions
Github Skills (15)
yolov410
configuration-management10
computer-vision10
pytorch10
image-segmentation10
3d-object-detection10
segmentation10
trainings10
python10
modeling10
realtime9
mask-rcnn9
faster-rcnn9
refactoring8
knockoutjs6
Programming languages (5)
Github contributions (5)
maudzung/Complex-YOLOv4-Pytorch
Jul 2020 - Jul 2022
The PyTorch Implementation based on YOLOv4 of the paper: "Complex-YOLO: Real-time 3D Object Detection on Point Clouds"
Role in this project:
Backend Developer
Backend DeveloperContributions:132 commits, 3 PRs, 53 pushes in 2 years
Contributions summary:Nguyen made several commits to the project, primarily focused on updating and modifying configuration files, most notably the `config.py` file. These changes included updates to data, model parameters, and training strategies such as learning rate scheduling. They also added functionality for integrating with the Kitti dataset, indicating work related to data processing and model configuration.
3d-object-detectionpoint-cloudpytorchyolov4object-detection
maudzung/TTNet-Real-time-Analysis-System-for-Table-Tennis-Pytorch
May 2020 - Nov 2020
Unofficial implementation of "TTNet: Real-time temporal and spatial video analysis of table tennis" (CVPR 2020)
Role in this project:ML Engineer
ML EngineerContributions:1 review, 257 commits, 1 PR in 6 months
Contributions summary:Nguyen implemented scripts for downloading, extracting, and preparing the dataset for training, which included downloading video files, extracting frames, and unzipping annotation files. They developed and revised scripts for the core preparation of data for training, including extracting selected frames based on event annotations and preparing datasets. Furthermore, they implemented the core TTNet model, including the ball detection, event spotting and segmentation modules, and added the functionalities for computing loss and training/validation phases.
event-detectiontabletennisspatialvideoreal-time
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.