Search AWS Educational Content

Learn how organizations can fine tune or re-train large language models to meet the needs of their unique use cases. By fine tuning an LLM with your own domain data you can provide the model with the vocabulary, context, and expertise needed to operate in your environment. Amazon SageMaker AI offers various model fine-tuning techniques such as supervised fine tuning, preference alignment, continued pre-training, and re-training to fine tune open source LLMs for your domain adapted LLMS.

Dive deep into IAM Access Analyzer in this interactive code talk focused on enhancing your AWS security posture. Watch and code along as we explore the automated reasoning powers of the three key analyzers: Unused Access, External Access, and Internal Access. Through live coding demonstrations, you'll learn how to configure each analyzer and interpret their findings.

Join leaders from ADP and Itau as they share how they reimagined their mainframe-based mission-critical systems. Dive deep into their transformation journey, exploring crucial architectural decisions, risk mitigation strategies, and lessons learned. Discover how they leveraged AWS Transform's automated code analysis, documentation, and business rules extraction, paired with other AWS developer tools, to achieve efficiency gains in reverse engineering, forward engineering, and testing.

Your AI agent calls a tool, gets an error, and instead of telling you, it generates a confident 'success' response. The user never sees the failure. Here are 5 techniques to stop that. Every time your agent responds, you're paying for tokens going in and tokens coming out. If what goes in isn't right, you're burning money and getting hallucinations back. I walk through 5 techniques to reduce token waste, improve acuracy, and catch failures before your users see them.

Legacy systems shouldn't hold back your AI ambitions. Join Dr. Asa Kalavade to see how AWS is making migration and modernization simpler and faster with agentic AI, building on nearly two decades of enterprise experience. Learn how we're using AI to build domain-specific agents that continuously learn, streamlining and accelerating transformations like never before.

Learn how organizations are modernizing their data architectures using the next generation of Amazon SageMaker to create AI-ready data platforms. This session covers how to design and build modern data architectures that enable both innovation and control, examining real-world patterns for data lake consolidation, cross-account governance, and AI integration.

In this episode of "AWS Show and Tell — What's New in Amazon Bedrock AgentCore Memory", we'll dive into the latest memory capabilities that make your AI agents smarter, more personalized, and truly context-aware across every interaction. Through hands-on demonstrations, you'll learn how to implement Episodic Memory to enable agents that reflect and learn across past conversations, leverage Streaming to read any new events from Long term memory, and apply proven memory patterns that combine...

The early days of Retrieval-Augmented Generation (RAG) are over. Naive RAG approaches (simple vector lookups fed to a LLM) are not enough to deliver accurate, scalable, and context-rich AI experiences. In this session you’ll learn how leading teams are pushing beyond basic document search to build agentic RAG pipelines that incorporate multi-step reasoning, dynamic namespace selection, metadata-aware retrieval, evaluation and more.

You can now host and run Claude Code, Codex, OpenCode, Kiro, or any coding agent on Amazon Bedrock AgentCore. Blog: https://go.aws/3QpFoo6 Code: https://go.aws/4vA2iIg

AI code generation is fast. Your ability to validate and ship it? Probably in need of an upgrade. Join Michael Webster, Principal Engineer at CircleCI, to learn about CircleCI’s AI engineering team’s experience deploying AI agents and coding assistants. Discover why 10x output is overwhelming traditional DevOps, which patterns unlock velocity without chaos, and why the solution isn't more tooling complexity, but smarter abstractions that make AI-generated code safer to ship.

Every Bundesliga match generates millions of data points, from every pass and sprint to every goal and unforgettable moment. But how do you turn all that data into real insights for coaches, players, and fans worldwide? Join Basil Fateen, AWS Principal Tech Evangelist, at the Franz Beckenbauer Supercup in Stuttgart as he explores how the Bundesliga is using AWS AI to reimagine the future of football.

GPU instances cost dollars per hour, not cents. Every second they're idle or spinning up is money lost. Here's how EKS Auto Mode gets AI workloads serving requests as fast as possible. Watch more: https://go.aws/4tjKcZH Try EKS Auto Mode: https://go.aws/4njDMbG #AWS #AmazonEKS #Kubernetes