Chanvichet V
Student
Cambodia
Summary
🤩
Rockstar🎓
Top SchoolChanvichet V is an aspiring software developer and MEng student at Tokyo Institute of Technology with eight years of hands-on experience across backend, MLOps, and DevOps. He contributes to high-profile open-source projects like ultralytics’ YOLOv5, where he optimized multi-GPU distributed training, improved logging, and hardened Windows workflows to streamline model training and deployment. His backend and infrastructure work on axolotl shows practical experience tuning training loops, callbacks, and optimizers to improve evaluation and operational reliability. A strong academic foundation—BS with a 4.0 GPA and international study in innovation leadership—complements his practical engineering skills. Based in Cambodia, he blends rigorous coursework with real-world contributions to ML tooling, signaling a fast-evolving engineer who moves quickly from research concepts to production-ready pipelines. He’s particularly effective at identifying and fixing subtle distributed-training issues that often block scalable ML workflows.
8 years of coding experience
Highschool, Highschool at East West International SchoolBachelor of Science - BS, Information Technology, 4.0/4.0, Bachelor of Science - BS, Information Technology, 4.0/4.0 at Sirindhorn International Institute of Technology (SIIT), Thammasat UniversityMaster of Engineering - MEng, Information and Communications Engineering, Master of Engineering - MEng, Information and Communications Engineering at Tokyo Institute of TechnologySoftware Development, Software Development at IT StepBachelor of Engineering - BE, Computer Software Engineering, Bachelor of Engineering - BE, Computer Software Engineering at Limkokwing University of Creative TechnologyExchange Program, Innovation Leadership, Exchange Program, Innovation Leadership at Czech Technical University in Prague
English, Khmer
Stackoverflow
196reputation
8kreached
14answers
2questions
Github Skills (26)
transformers10
pytorch10
back-end-development10
python10
machine-learning10
deeplearning-ai10
mlops10
deep-learning10
object-detection10
backend10
optimizers9
dockers9
optimizer9
docker9
devops9
Programming languages (14)
Github contributions (5)
axolotl-ai-cloud/axolotl
May 2023 - Apr 2025
Go ahead and axolotl questions
Role in this project:
Backend Developer & DevOps Engineer
Backend Developer & DevOps EngineerContributions:483 reviews, 427 PRs, 434 pushes in 1 year 11 months
Contributions summary:Chanvichet primarily focused on enhancing the backend and infrastructure of the project. Their contributions included adding evaluation batch size, integrating callbacks for saving PEFT models, and incorporating additional callbacks into the trainer. Furthermore, the user modified the learning rate scheduler and optimizer configurations. These changes indicate efforts to improve model evaluation, training efficiency, and overall operational aspects of the project.
ultralytics/yolov5
Dec 2020 - Apr 2021
YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite
Role in this project:MLOps Engineer
MLOps EngineerContributions:6 reviews, 5 commits, 23 PRs in 4 months
Contributions summary:Chanvichet primarily focused on optimizing the training and deployment process of the YOLOv5 model. Their contributions include fixing DDP training issues related to multi-GPU setups, local rank arguments, and EMA attribute errors. They also improved logging, added an InfiniteDataLoader class, and addressed Windows-specific download issues, contributing to a more robust and efficient training pipeline.
pytorchdeep-learningyolov8object-detectionmachine-learning
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.