Presenter

Joey Conway is a Director of Deep Learning Software Product Management at NVIDIA and for the past 4 years, he has been working on and supporting NVIDIA's contributions to TensorFlow, PyTorch, Triton Inference Server, Riva for Conversational AI, and Merlin for Recommendation Systems. Most recently, he has started helping with productizing research from Megatron and large scale NLP models for training and inference deployment. He graduated from MIT Sloan with a Master's degree in Business Administration. Joey’s expertise in deep learning software make him a natural and which brings him to Supercomputing ’21 to contribute to the panel discussion with a slant towards NVIDIA’s technology perspective. He’ll share general guidance on the challenges and successes of creating large scale language models.