Domino at NVIDIA GTC Replay

TopDanmark’s Head of Machine Learning, Stig Pederson, and Domino CEO, Nick Elprin, discuss how Topdanmark, one of Denmark’s largest insurers, is infusing models throughout operations for fraud detection, policy approvals, and risk management - improving customer experience for a competitive advantage using leading-edge technology like Natural Language Processing (NLP). They’ll discuss what it takes to become model-driven by removing process and technology bottlenecks for unobstructed data science innovation.

The convergence of analytics and AI workloads with simulation or classical High Performance Computing (HPC) workloads in the cloud are becoming requirements for the model-driven businesses of the future. For many companies today, these siloed environments limit performance at the production scale necessary for truly transformative data science innovation. In this session, Dr. Revati Kulkarni, Technology Head of the HPC and AI group at TCS, along with Thomas Robinson, VP of Strategic Partnerships at Domino, discuss role-based use cases they’ve seen in enterprise customers. Learn best practices for building a heterogeneous platform to unite analytics and AI workloads, backed by the HPC technology for production level workflows.

Apache Spark has been the incumbent distributed compute framework for the past 10+ years. But the overhead and complexity of Spark have been eclipsed by new frameworks like Dask and Ray. In this talk we will discuss trade offs of these frameworks, explore choosing the appropriate framework and demonstrate the ease of using Dask or Ray. Additionally, we'll present an integrated solution based on the Domino Data Science Platform, and NVIDIA NGC containers, which enables data scientists to easily provision a Spark/Ray/Dask cluster with an arbitrary number of GPU-accelerated workers, and access it through their favorite integrated development environment.

Domino's own Chief Data Scientist, Josh Poduska, joins a panel of NVIDIA DGX-certified partners to discuss best practices, trends, and recommendations for managing data science teams, data, and optimizing and accelerating data science workloads and running heavy training workloads.