Bryan Catanzaro
VP of Applied Deep Learning Research at NVIDIA; co-author of the Megatron-LM and cuDNN papers
Bryan Catanzaro is Vice President of Applied Deep Learning Research at NVIDIA, where Catanzaro leads a team applying deep learning to problems that range from language understanding and video games to computer graphics and chip design. Catanzaro co-authored the 2014 paper introducing cuDNN, a library of optimized deep learning routines for GPUs, and the 2019 Megatron-LM paper on training multi-billion-parameter language models with model parallelism; NVIDIA's developer blog also credits Catanzaro with helping lead the team that invented DLSS 2.0. Catanzaro appears in the leadership credits of NVIDIA's Nemotron 3 white paper and wrote NVIDIA's September 2025 blog post on how Nemotron models, datasets and techniques are shared openly. Before NVIDIA, Catanzaro worked at Baidu on systems for training and deploying end-to-end deep learning speech recognition.123457
Work
Megatron-LMOpen-source code
Co-author, with Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley and Jared Casper, of the 2019 Megatron-LM paper on intra-layer model parallelism for training transformer language models; the NVIDIA/Megatron-LM repository README asks users to cite this paper.56
NVIDIA NemotronOpen weights
Listed in the leadership credits and the NVFP4 contributor group of the NVIDIA Nemotron 3 white paper (December 2025), which states that NVIDIA will release the model weights, training datasets and recipes; also wrote NVIDIA's September 2025 post "Open Secret: How NVIDIA Nemotron Models, Datasets and Techniques Fuel AI Development".71
cuDNN (external site: arxiv.org)Published research
Co-author of "cuDNN: Efficient Primitives for Deep Learning" (2014), which describes a BLAS-like library of optimized deep learning routines for GPUs; NVIDIA's developer blog says Catanzaro's research at NVIDIA led to the creation of cuDNN.42
Other documented roles
Roles the sources document but that are not confirmed as current.
- Worked on systems for training and deploying end-to-end deep learning speech recognition, Baidu2
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.