Vinh Nguyen
Deep Learning Engineer And Data Scientist
Melbourne, Victoria, Australia
Summary
👤
Senior🎓
Top SchoolVinh Nguyen is a Deep Learning Engineer and Data Scientist with over a decade of experience bridging academic research and production AI, currently applying GPU-accelerated deep learning at NVIDIA. He holds a PhD in Machine Learning, has authored 50+ papers with 4,500+ citations and two best paper awards, and secured over A$400k in competitive research funding. Vinh is a Kaggle Competition Master and contributor to prominent open-source projects—especially TensorRT integrations for PyTorch and TensorFlow and mixed-precision training improvements—demonstrating expertise in model optimization for real-world inference. His background spans recommender systems, speech recognition, LLMs and NLP, and he combines rigorous research skills with hands-on engineering to take models from experimentation to reproducible deployment. A less obvious strength is his consistent focus on performance tuning (AMP, INT8 calibration, TensorRT) that directly accelerates production workloads.
10 years of coding experience
8 years of employment as a software developer
PhD, Data Mining, Machine Learning, Bioinformatics, PhD, Data Mining, Machine Learning, Bioinformatics at University of New South WalesPhD Intern, Data Mining, PhD Intern, Data Mining at National Institute of Informatics, Tokyo, JapanBE (Hons), Information Technology, specialized in software engineering and grid computing, BE (Hons), Information Technology, specialized in software engineering and grid computing at Hanoi University of Technology
French, Vietnamese, English
Github Skills (31)
pytorch10
python10
tensorrt10
machine-learning10
keras10
deeplearning-ai10
deep-learning10
tensorflow10
gpu10
optimisation10
natural-language-processing10
recommender-system10
cuda10
computer-vision10
jupyter-notebook10
Programming languages (7)
Github contributions (5)
tensorflow/tensorrt
Jun 2019 - Apr 2022
TensorFlow/TensorRT integration
Role in this project:
ML Engineer
ML EngineerContributions:3 reviews, 14 commits, 18 PRs in 2 years 10 months
Contributions summary:Vinh added a Jupyter notebook example demonstrating how to create and use a TF-TRT (TensorFlow-TensorRT) optimized model from a saved TensorFlow model for image classification. The notebook includes steps for verifying the original FP32 model, creating TF-TRT FP32, FP16, and INT8 models, and calibrating the INT8 model using ImageNet validation data. This demonstrates the user's focus on optimizing TensorFlow models for inference using TensorRT.
nvidia-dockerdeep-learningtensorflowtensorrt
pytorch/TensorRT
Jun 2020 - Jul 2022
PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT
Role in this project:ML Engineer
ML EngineerContributions:19 commits, 4 PRs, 15 comments in 2 years
Contributions summary:Vinh's contributions focused on adding demo notebooks to the repository, specifically showcasing the integration of PyTorch models with NVIDIA's TensorRT. Their work included the creation of notebooks demonstrating object detection using SSD models and deep learning model compilation with trtorch. They also addressed input dimension issues and provided a Dockerfile for reproducibility.
cudapytorchjetsonnvidiadeep-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.