Predibase: A Low-Code Deep Learning Platform Built for Scale

Explore a groundbreaking low-code machine learning platform in this Ray Summit 2022 conference talk. Dive into Predibase, created by the minds behind Ludwig and Horovod, which combines large-scale deep learning with cutting-edge model architectures for NLP, computer vision, tabular, and hybrid data. Learn how Predibase differentiates itself from traditional AutoML solutions by offering a modular "lego brick" approach, enabling both data scientists and engineers to iteratively construct optimal models without mastering complex frameworks like PyTorch or TensorFlow. Discover how Predibase leverages Ludwig and Ray to deliver a comprehensive, cloud-native solution that covers data processing, model training, hyperparameter optimization, and model serving at scale. Gain insights into Predibase's unique serverless abstraction over Ray, which facilitates multi-cloud and multi-tenant deployments capable of handling numerous concurrent users and requests.