Summary
Tuan Le Duc Anh is a Senior AI Engineer based in Hanoi with 5 years of hands-on experience building high-performance ML systems from research prototypes to production. He has delivered state-of-the-art results across speech, vision, and video—authored MedNeXt for medical segmentation, designed 3DNeRV for efficient video compression, and led ASR and speech-tokenizer work that achieved SOTA on regional benchmarks and an Interspeech paper. At Moreh he engineered a pure HIP C++ stack for GPT inference on AMD MI250s, optimizing multi-GPU communication, FlashAttention and GEMM to hit impressive single-node throughput for very large models. Comfortable bridging research and engineering, he’s led teams and community events (GDG Hanoi) and packaged modular tools like Ichigo on PyPI to help platform developers adopt async audio pipelines. An unusual strength is shipping both low-level GPU kernels and high-level deployed services, making him effective across the full ML stack.
4 years of coding experience
3 years of employment as a software developer
Bachelor of Science - BS, Data Science and Artificial Intelligence, Bachelor of Science - BS, Data Science and Artificial Intelligence at Hanoi University of Science and Technology
English