Karthik Kumar Odapally

AI training failures at scale cost time and money. This session demonstrates how to build automated error detection and remediation into containerized AI environments using Amazon EKS for both Trainium and NVIDIA GPUs.

Riot Games ingests about 20 TB of data every day on AWS. This data powers a wide range of services including game matchmaking, in-game personalization, analytics, security, and player behavior management.

As Amazon’s consumer business continues to grow, so does the volume of data and complexity of analytics done to support the business. In this session, hear how Amazon.com uses AWS technologies to build a secure, compliant environment to process and analyze petabytes of data.

One thing we’ve learned is that failures come in many forms, some expected and some unexpected. A core consideration is how to minimize the blast radius of any failures.

Amazon Prime Video is one of the largest streaming services in the world, serving over 8 percent of all internet traffic in the US on any day. With millions of devices in over 200 countries around the world, collecting and analyzing telemetry data in real time is a major scaling challenge.

Woot.com designed and developed a data lake as a replacement for their legacy data warehouse to deliver powerful analytics capabilities across multiple business areas. In this session, learn how it used Amazon EMR, Amazon Athena, AWS Glue, and Amazon QuickSight to build an automated process to ingest and centralize data from external and internal sources for immediate analysis.

This presentation was recorded prior to re:Invent. In this session, learn how Zappos developed an end-to-end machine learning (ML) platform leveraging AWS analytics.

In this session, learn how an analytics platform team in Amazon Prime Video processes, governs, manages schemas, and stores data at petabyte scale. Discover how they were also able to secure and deliver easy access to teams across the Prime Video organization in a heterogeneous technology environment.