Search AWS Educational Content

Context engineering is one of the most important tools we have for building efficient AI agents. It's about being intentional with what information makes it into an agent's context window, what stays there, what gets removed or transformed, and how the agent gets the information it needs when it needs it.

AWS Glue is a serverless data integration service that makes it easy to discover, prepare, and combine data for analytics, machine learning, and application development. In this video learn, how Glue with Apache Spark 4.1 runtime supports VARIANT shredding on Apache Iceberg v3 data type to manage semi-structured data. And with Declarative Pipelines, you simply define what you want, the engine handles execution order, dependencies, and recovery automatically.

Twitch stream from aws

Building an AI agent that works reliably without burning through your token budget takes a lot more engineering than most demos make obvious. In this video, we're breaking down Strands harness, a new general-purpose agent harness from AWS that packages context management, memory, tools, skills, prompt caching, and more into tuned defaults you can use out of the box. The launch benchmarks report state-of-the-art agent performance with 28% lower token cost.

🛠️ Spec Driven Development Best Practices with Kiro | How to Build Like AWS | E4 Are you tired of vibe coding your way into a mess? In this episode of **How to Build Like AWS**, we dig into one of the most powerful shifts in AI-assisted development: **spec-driven development with Kiro**. The AWS PACE team joins us to break down how they build real customer prototypes — fast — and why having a spec isn't just good practice, it's a competitive advantage.

Watch Ben Gilbert and David Rosenthal, co-hosts of Acquired—the #1 technology show on Spotify and Apple Podcasts—take the stage with special guests including Max Neukirchen, Global Co-Head, J.P. Morgan Payments, Greg Peters, Co-CEO, Netflix, Aravind Srinivas, Co-founder and CEO, Perplexity, and Matt Garman, AWS CEO.

In this video, learn how Indeed adopted Amazon S3 Tables with Apache Iceberg to eliminate data silos, reduce infrastructure overhead, and unlock data access across the organization so every team can draw better insights and help job seekers and employers find matches.

In Formula 1, victories depend on turning millions of data points into winning decisions. Organizations can apply this same precision to unlock value from underutilized data assets. Drawing from F1's model, we present a three-part framework for data excellence: customer-focused decisions, dynamic people- powered strategies, and market-reive structures. Learn practical approaches to evolve your data strategy into a business advantage in the AI era. Learn more about AWS events: https://go.

Watch Matt Wood, Chief AI & Technology Officer at AWS, recap the biggest announcements from the AWS New York Summit 2026 keynote in just a few minutes. This year's keynote introduced a simple idea: agents create compounding momentum. The more you use them, the better they get. The better they get, the more you trust them. And the more you trust them, the more you hand off.

Many enterprises struggle to move agentic AI beyond proof-of-concept. Discover how Amazon Bedrock AgentCore and Claude are making it easier to deploy production-ready AI agents at scale. Learn more: 📖 Read the full blog post: https://go.aws/4rmUcSj 🔧 Get started with Amazon Bedrock AgentCore: https://go.aws/4nJmvrP 🤖 Explore Claude Sonnet 4.5: https://go.aws/47o75lQ Subscribe to AWS: https://go.aws/subscribe Create a free AWS account: https://go.aws/signup Try AWS for free: https://go.

Vibe coding with AI to write functions & fix bugs? Table stakes. Building an agent that architects, implements, tests, and deploys a full application to AWS? That's fundamentally different. This session explores design patterns that enable Claude on AWS Bedrock to achieve "escape velocity"—the ability to make sustained, autonomous progress toward production deployment. You'll learn Anthropic's core principles for building agents that can handle full-stack development tasks independently.

Learn how to build, deploy and operate agents securely at scale - using any framework and model. Dennis Traub, Developer Advocate, joins Gunnar Grosch to explain how AgentCore's building blocks can help you get agents into production. See how to deploy agents using multiple frameworks. Find out how to turn an existing AWS Lambda function behind an Amazon API Gateway endpoint into a helpful agentic tool using MCP. We are also available to answer any of your serverless questions.