Search AWS Educational Content

High performance computing workloads are growing in complexity, scale, and importance. AWS Parallel Computing Service (PCS) brings simplicity and flexibility to this challenge by automating cluster creation, job scheduling, and resource management. In this session, learn how PCS helps organizations move faster, scale securely, and achieve predictable performance for demanding workloads in energy, manufacturing, and research.

A multiplayer game server that spins up when a player joins, holds state the entire session, suspends when idle (billing stops), and resumes in under a second. One VM per room, zero servers to manage, built on Lambda MicroVMs. Lambda MicroVMs are a new compute primitive in the Lambda family. Same Firecracker isolation that already runs billions of Lambda invocations, but a different shape: stateful, long-running (up to 8 hours), session-oriented, with suspend/resume and per-second billing.

Feature flags have evolved beyond simple on/off switches for software releases. This code-focused session demonstrates advanced feature flagging techniques through real implementation examples and production code samples. See how Amazon leverages feature flags at scale using AWS AppConfig, with detailed code walkthroughs that showcase powerful use cases beyond basic feature releases.

AWS Graviton-based Amazon EC2 instances provide the best price performance for workloads in Amazon EC2. In this session, dive deep into the AWS Graviton processor and learn about its workload performance, energy efficiency, and software offerings. Graviton5 processor, our newest Graviton processors. Learn about common use cases, best practices to optimize your workloads across various applications, customer success stories, and how you can accelerate your AWS Graviton journey.

Kiro is an AI-native development environment that transforms how developers build applications. We've covered Kiro last September on the show and now we're checking in 6 months later to see how its evolved. Today we'll teach Kiro about our codebase, develop some new features along with the chat, and dive into advanced features like agent hooks, MCPs, and sub-agents. You can try out Kiro for yourself at https://kiro.dev/!

Enterprises use Apache Kafka and Apache Flink for an increasing number of mission-critical use-cases, real-time analytics, application messaging, and machine learning. As this usage grows in size and scale, so does the criticality, scale, and cost of managing the Kafka and Flink clusters.

Join us to learn about Apache Iceberg Version 3 (V3) and how it addresses critical challenges in modern data lakes. In this session, we'll explore how V3's variant data type, deletion vectors and row lineage capabilities help organizations reduce storage costs, accelerate data operations, and maintain comprehensive audit trails at petabyte scale. We'll also look ahead to what's coming in Apache Iceberg V4.

Organizations struggle to customize AI models for complex tasks due to the extensive human annotation required for traditional fine-tuning approaches. Learn how Amazon Bedrock's new Reinforcement Fine Tuning (RFT) capabilities enable developers to train models using custom evaluators instead of large labeled datasets. Discover how to implement RFT with Amazon Nova and open-source models using techniques like RLVR for code generation and RLAIF for AI feedback.

What does it take to bring world-class connectivity to the billions of people and businesses that have never had reliable access? Tom Soderstrom, AWS Executive in Residence, and Chris Weber, VP of Business at Amazon Leo, share how Amazon Leo (formerly known as Project Kuiper) is building toward that vision one launch at a time, with over 200 satellites in orbit since this conversation took place in December 2025.

Meet the future of customer experience: AI agents that understand, reason, and take action. Amazon Connect Customer AI agents dynamically navigate your organization’s resources to find solutions and take action to resolve customer needs. They handle many issues on their own, but also work in collaboration with your workforce to deliver personal, effortless customer experiences. Subscribe to AWS: https://go.aws/subscribe Create a free AWS account: https://go.

Physics AI models for automotive aerodynamics traditionally struggle with real-world applications due to limited training data diversity. PhysicsX addresses this challenge with their new Large Physics Model (LPM), powered by AWS's GPU-accelerated CFD simulations and active learning to generate diverse automotive geometry training data.

Enterprise-grade AI needs a unified data strategy for accurate, reliable results. Learn how knowledge graphs bridge the gap between how the real world operates and how companies actually store and organize data - transforming data into knowledge. See how Agentic GraphRAG (retrieval-augmented generation with knowledge graphs) drives real success: Learn how companies like AbbVie have deployed GenAI to build chatbots grounded in knowledge graphs, improving productivity and trust, while a major...