Dmitri Vainbrand

Dmitri Vainbrand is a senior deep learning architect at GPU DL Architecture group at NVIDIA. Dmitri is working on GPU HW and system architecture, focusing on large deep learning networks, systems, and scale-out. Dmitri received his MSc from Technion in 2011 in the field of network-on-chip architectures for neural networks. Prior to NVIDIA, he worked at Intel, where he took part in the development of five generations of CPU Core processors and Intel’s first AI accelerator.
Avatar photo

Posts by Dmitri Vainbrand

Conversational AI

Scaling Language Model Training to a Trillion Parameters Using Megatron

Natural Language Processing (NLP) has seen rapid progress in recent years as computation at scale has become more available and datasets have become larger. At... 17 MIN READ