Supersonic Model Serving Using Apache Licensed Deep Java Library (DJL) and Quarkus

Discover how to implement supersonic model serving in a cloud-native environment using Apache Licensed Deep Java Library (DJL) and Quarkus in this informative conference talk. Learn about the challenges data scientists face when deploying trained models to production and how Amazon's DJL project addresses the skill-set mismatch between data scientists and enterprise DevOps teams. Explore the benefits of using DJL, which allows data scientists to create and refine models using their preferred tools while enabling enterprise DevOps teams to serve these models on Java Virtual Machines. Gain insights into Red Hat's Build of Quarkus, a supersonic Java framework designed for modern cloud-native architected business applications. Understand how the combination of DJL and Quarkus provides a powerful solution for serving large models requiring parallelism across GPUs in enterprise environments. This talk offers valuable knowledge for data scientists, DevOps professionals, and anyone interested in bridging the gap between model development and production deployment in cloud-native ecosystems.