yt:channel:T-nPlVzJI-ccQXlxjSvJmwT-nPlVzJI-ccQXlxjSvJmwAWS DevelopersAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2014-01-09T20:39:19+00:00yt:video:cKzYBRE0jI4cKzYBRE0jI4UCT-nPlVzJI-ccQXlxjSvJmwHackathon Tips for BeginnersAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-21T18:30:46+00:002026-08-22T08:51:07+00:00Hackathon Tips for BeginnersFirst hackathon? We asked hackathon judges at WeAreDevelopers Congress Berlin for their best advice for first-time participants.
These tips apply whether you're building solo or with a team. And if you're looking for a place to start, the Agents for Humans Hackathon is live now on DevPost. Build an AI agent that improves everyday life using the Strands Agents SDK.yt:video:KlBO0653BQAKlBO0653BQAUCT-nPlVzJI-ccQXlxjSvJmwBuilding Apps on AWS Using Claude CodeAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-20T21:00:41+00:002026-08-22T16:06:52+00:00Building Apps on AWS Using Claude CodeTwo environment variables to connect Claude Code to Amazon Bedrock. Model calls never leave AWS, everything bills to your account. I install the Agent Toolkit, build an agent with Strands, and show the same setup running in Claude Desktop.
Two ways to connect: raw environment variables (two lines, works immediately) or `claude setup bedrock` (persists to settings.json, survives restarts). I show the exact commands for both.
The Agent Toolkit for AWS is a plugin that gives Claude Code best practices and skills for building on AWS services. Once installed, you get /agents slash commands that scaffold Strands agents, wire up tools, and generate working code. I use it to build a flight-tracking and weather agent, test it on the local dev server, and confirm it works.
Agent Toolkit for AWS: https://go.aws/3So42q5
Read the blog: https://go.aws/4galyHH
Claude Code on Amazon Bedrock Docs: https://go.aws/4zpUnjy
Follow AWS Developers!
🆇 X: https://go.aws/4zuSLVA
💼 LinkedIn: https://go.aws/3U8Ko23
0:00 Claude Code on AWS: Two Env Variables
1:33 Method 1: Environment Variables
2:10 Method 2: Claude setup Bedrock
4:29 Install the Agent Toolkit for AWS
5:17 Building a Flight & Weather Agent
6:43 Demo: Agent Running Locally
8:00 Claude Desktop + Bedrock Setup
8:58 Configuring Claude Desktop for Bedrock
12:06 Recap
#AWS #AgentToolkit #AmazonBedrockyt:video:WcAksz0qPcQWcAksz0qPcQUCT-nPlVzJI-ccQXlxjSvJmwBuilding Multiplayer Game Servers with AWS Lambda MicroVMsAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-18T19:00:15+00:002026-08-20T12:04:52+00:00Building Multiplayer Game Servers with AWS Lambda MicroVMsA 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.
I built Faceteroids (asteroids but you shoot my face) to show how the whole thing works. One microVM per game room, WebSocket connections, server-authoritative physics, and compute billing that stops the instant a room goes idle.
Read the Blog: https://go.aws/4goEwZZ
Github Repo: https://go.aws/4gga8Ro
Lambda MicroVMs Docs: https://go.aws/4g4jrFf
Follow AWS Developers!
🆇 X: https://go.aws/4glsbWv
💼 LinkedIn: https://go.aws/4zqVcIW
0:00 Building a Multiplayer Game Server on Lambda
1:59 Lambda Functions vs Lambda MicroVMs (Side-by-Side)
3:05 Faceteroids: The Architecture
4:08 Three Layers: Control Plane, Frontend, Game Server
6:08 MicroVM Lifecycle: Image to Termination
11:07 Ready Hook & Validate Hook
12:21 The Snapshot Trap (RNG Seed Bug)
14:10 3 Rules When Planning Snapshots
19:11 Browser WebSocket Auth (Sub-Protocol Trick)
21:02 Pricing: Per-Second Billing & Burst Scaling
24:51 Beyond Games: Other Use Cases
#AWS #LambdaMicroVM #Serverlessyt:video:X-nAoDAjWsoX-nAoDAjWsoUCT-nPlVzJI-ccQXlxjSvJmwRun Coding Agents Remotely on Amazon Bedrock AgentCoreAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-14T17:00:25+00:002026-08-14T17:00:25+00:00Run Coding Agents Remotely on Amazon Bedrock AgentCoreWe're all tired of leaving laptops half open while the coding agent runs. These agents need a shell, a file system, credentials, and your laptop won that job by being nearby, not by being safe. An agent importing an unvalidated dependency or overwriting config files on your local machine is a real risk. Amazon Bedrock AgentCore gives each agent its own isolated environment in the cloud.
AWS Blog Post: https://go.aws/45mTwCB
Github Repo: https://go.aws/4xA5TXS
Follow AWS Developers!
🆇 X: https://go.aws/4wZaBi3
💼 LinkedIn: https://go.aws/4y0eweL
0:00 The Problem: Laptop as Agent Runtime
0:20 What Coding Agents Actually Need
0:35 4 Agents, 1 Bug, Isolated Environments
0:46 The Bug: Delete Filter Flipped
1:38 Close the Laptop, AgentCore Handles It
#AWS #AmazonBedrock #AgentCoreyt:video:QppcKeWmSl0QppcKeWmSl0UCT-nPlVzJI-ccQXlxjSvJmwSystem Design for BeginnersAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-12T19:01:03+00:002026-08-16T03:14:57+00:00System Design for BeginnersSystem design interviews aren't just about knowing the right answer; they are about how you think. Here are 3 tips to help you approach any system design problem with confidence: start with the problem (not the technology), think critically about every decision, and invest in your foundations.
Watch the full video on how to break down a classic system design interview question step by step: https://go.aws/3UarF64yt:video:_VXcDKI7JvA_VXcDKI7JvAUCT-nPlVzJI-ccQXlxjSvJmwMCP: When to Use It (And When You Shouldn't)AWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-11T17:15:10+00:002026-08-15T17:57:09+00:00MCP: When to Use It (And When You Shouldn't)Everyone says "just wrap it in MCP." And for reusable, shared, remote tools, they're right. But every MCP capability sits in the context window every turn, whether the model uses it or not. That adds up, and sometimes a plain function call is all you need.
MCP (Model Context Protocol) explained with a live demo and an honest take on the trade-offs. We cover what problem MCP solves (the N-agents x M-tools integration explosion), how the protocol actually works (client, server, 7-step flow), and a before/after demo showing an agent answering from stale training data vs pulling live docs through an MCP server.
Agent Toolkit: https://go.aws/4wEs7Hy
AWS Blogpost: https://go.aws/4zguYce
Follow AWS Developers!
🆇 X: https://go.aws/4z55PAV
💼 LinkedIn: https://go.aws/4bKDdTI
0:00 MCP: Sometimes the Wrong Tool for the Job
0:34 The Problem (5 Agents x 10 Tools = 50 Integrations)
1:27 How MCP Works (USB-C for AI Agents)
3:25 Demo: Without MCP (Outdated Answer)
4:08 Demo: With MCP (Live Docs)
5:36 The Trade-Off: Context Window Cost
6:29 When to Use MCP vs Lighter Alternatives
7:29 Recap
#AWS #MCP #AIAgentsyt:video:7alMWE6vo3Y7alMWE6vo3YUCT-nPlVzJI-ccQXlxjSvJmwRunning Claude Code on Amazon BedrockAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-06T19:30:54+00:002026-08-10T01:07:52+00:00Running Claude Code on Amazon BedrockClaude Code is an agentic coding tool by Anthropic that helps you build features, fix bugs, and automate development tasks. But some developers can't use it due to data residency, compliance, or billing requirements. With Amazon Bedrock, you can run Claude Code with your data staying entirely within AWS — no data leaving your infrastructure.
We'll give you the complete walkthrough: environment setup, cross-region inference, prompt caching, the new Amazon Bedrock Mantle endpoint, per-team cost tracking, and Amazon Bedrock Guardrails for content governance.
#Claude #ClaudeCode #AWS #AmazonBedrockyt:video:sP4CNT3pDsMsP4CNT3pDsMUCT-nPlVzJI-ccQXlxjSvJmwLangGraph for BeginnersAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-06T17:30:10+00:002026-08-10T01:04:38+00:00LangGraph for Beginners"Can you build agents?" is one of the most in-demand lines in job posts right now. LangGraph is the open-source framework behind it. Zero to a working support agent with routing, RAG, tools, and human-in-the-loop, powered by Amazon Bedrock.
LangGraph explained from scratch, then built into a working customer support agent for a hair care company. The app classifies intent, routes to a specialist (product advisor, returns agent, or general responder), retrieves from a Bedrock Knowledge Base using RAG, and pauses for human approval before issuing refunds.
AWS Blogpost with Resources: https://go.aws/4595aRs
AWS Agent Toolkit: https://go.aws/4xkW7sz
Follow AWS Developers!
🆇 X: https://go.aws/4xrr1zG
💼 LinkedIn: https://go.aws/4bym7rZ
0:00 Why LangGraph Matters Right Now
1:15 When Does a Business Need This?
2:40 Context Overload: Why One Agent Can't Do Everything
3:26 How to Think in LangGraph (Draw the Flowchart)
4:08 Core Concepts: Nodes, Edges & State
5:16 When to Cut a New Node (2 Questions)
6:12 Building a Flowchart: Hair Care Support Bot
7:37 Three Specialist Lanes + RAG
9:35 Human in the Loop: Refund Approval Gate
11:08 Live Demo & Code Walkthrough
27:09 Demo: Product Question Flow
29:53 Demo: Refund Flow + Human Approval
35:19 Your Turn: Build It with Kiro + AWS Agent Toolkit
#AWS #LangGraph #AmazonBedrockyt:video:xh10ChpIn3Ixh10ChpIn3IUCT-nPlVzJI-ccQXlxjSvJmwWhat Happens When You Call Amazon Bedrock?AWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-04T22:45:17+00:002026-08-09T01:04:42+00:00What Happens When You Call Amazon Bedrock?You call Bedrock, words come back, you ship the feature. But that gap between your request and the response is where the confusing stuff lives: why streaming feels faster, why answers cut off, why temperature changes output. All of it comes from one mechanism.
One request, followed from your code all the way through to the words coming back. The full pipeline: auth and routing, tokenization, attention, the probability distribution the model builds for every single word, softmax, temperature, top p, the autoregressive loop, and streaming.
Agent Toolkit for AWS: https://go.aws/4w2ip17
Amazon Bedrock: https://go.aws/4whL0jm
Follow AWS Developers!
🆇 X: https://go.aws/4wGKXyt
💼 LinkedIn: https://go.aws/3TxUt8s
0:00 One Trick: What Word Comes Next
0:39 What Happens When You Call Bedrock
1:57 The Model Isn't Learning from You
2:55 Auth & Routing (Before the Model Runs)
4:00 Tokenization: Text to Numbers
5:05 Reading the Prompt (Attention)
5:49 Picking the Next Word (The Real Math)
7:21 Temperature Explained
8:08 Top P Explained
8:36 The Autoregressive Loop: Pick, Append, Repeat
9:04 The Math Explained
10:03 Tokens Back to Text + Streaming
10:50 Why Everything Makes Sense Now
12:12 Your Turn + Agent Toolkit for AWS
#AWS #AmazonBedrock #AgentToolkityt:video:Uuuked8_I50Uuuked8_I50UCT-nPlVzJI-ccQXlxjSvJmwBuild & Deploy a Java AI Agent in Under 3 MinutesAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-08-04T17:14:55+00:002026-08-09T01:05:56+00:00Build & Deploy a Java AI Agent in Under 3 MinutesJava developers: Build and deploy an AI agent on AWS in under 3 minutes.
#SpringAI #AgentCore #JavaAI #AWS #BedrockAgents #Kiroyt:video:H-woAjsLcuEH-woAjsLcuEUCT-nPlVzJI-ccQXlxjSvJmwWhat is an Agent Harness? (And How We Built One)AWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-07-29T19:43:55+00:002026-08-03T05:39:18+00:00What is an Agent Harness? (And How We Built One)Getting an AI agent working on your laptop is the easy part. Hosting it for real users, scaling it, keeping it healthy when more than one person shows up. That's three layers deep: Bedrock for the brain, Strands for the harness, AgentCore for production.
The model is the brain. The harness is the hands, infrastructure, and memory that let it do real work. You define tools, write a prompt, the model figures out what to call and when.
GitHub: https://go.aws/4fvfmd2
#StrandsAgents #AgentHarness #AWSyt:video:BMI8GXmOPw8BMI8GXmOPw8UCT-nPlVzJI-ccQXlxjSvJmwContainers on Amazon ECS with Mama JAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-07-28T22:14:19+00:002026-08-03T05:40:40+00:00Containers on Amazon ECS with Mama JMama J is back, and this time it's Containers.
AWS Developer Advocate Eric Johnson takes on the classic "but it works on my machine" problem and explains containers with his mom, Mama J, who still hasn't touched a line of code.
Watch the full video here → https://go.aws/4vVSgB0yt:video:Ah8rrgOq-c8Ah8rrgOq-c8UCT-nPlVzJI-ccQXlxjSvJmwThe Only Networking You Need to Start Building on AWSAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-07-23T16:17:09+00:002026-07-23T16:17:09+00:00The Only Networking You Need to Start Building on AWSYou don't need deep networking knowledge to start building on AWS. You need one mental model and about 10 minutes. The neighborhood analogy covers VPC, public vs private, security groups, routing, and the default setup that's already there when you create an account.
AWS networking explained through a single analogy: a neighborhood. VPC is the gated community. Houses are your resources. Public streets connect to the highway (internet). Interior streets are private. Security groups are the guard at each door. Routing is whether there's a road connecting point A to point B.
Blog post: https://go.aws/4pSH7jj
AWS Networking Essentials: https://go.aws/4fC0JU6
Follow AWS Developers!
🆇 X: https://go.aws/4pHfKbO
💼 LinkedIn: https://go.aws/4gS1uus
0:00 You Don't Need Deep Networking Knowledge
0:50 The Only Question: Who Can Talk to What?
1:56 Public vs Private Resources
3:20 Security Groups = Guard at the Door
4:22 Routing, Internet Gateway & NAT Gateway
5:40 Default VPC: Already Set Up for You
8:51 Full Request Flow: Internet to App to Database
9:42 Troubleshooting: Common Networking Problems
10:25 Conclusion
#AWS #Networking #CloudComputingyt:video:igsklrIO3foigsklrIO3foUCT-nPlVzJI-ccQXlxjSvJmwBuild AI Agents on AWS in Under 10 Minutes (Bedrock, Strands, AgentCore)AWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-07-21T13:57:42+00:002026-07-25T06:23:44+00:00Build AI Agents on AWS in Under 10 Minutes (Bedrock, Strands, AgentCore)Getting an AI agent working on your laptop is the easy part. Hosting it for real users, scaling it, keeping it healthy when more than one person shows up. That's three layers deep: Bedrock for the brain, Strands for the harness, AgentCore for production.
Three AWS building blocks that take an AI agent from your laptop to production, explained with running code.
Bedrock is the brain. One managed service with a menu of models (such as Anthropic, OpenAI, open-weight and more). No servers, no GPUs. You call a model and get a response.
Strands is the harness. An open source SDK that wraps the model with tools and a loop. The model reasons, picks a tool, Strands runs it, the result comes back, and the model decides what's next. That cycle is the agent loop, and you write none of the orchestration code.
AgentCore is the production environment. A managed, serverless runtime for agents. You containerize with the CLI, deploy, and get an endpoint. No web server to write, no scaling to configure. It also brings extras you pull in when needed: memory (agent remembers across sessions), gateway (existing APIs become agent tools), observability (traces of what the agent did and why) and more.
Blog post (full tutorial): https://go.aws/4ptSbTC
Agent Toolkit: https://go.aws/4wNkEGn
Follow AWS Developers!
🆇 X: https://go.aws/3THWG0T
💼 LinkedIn: https://go.aws/4fHqtzn
0:00 Laptop to Production: The Real Problem
0:20 Amazon Bedrock: The Brain
0:52 Code: Calling Bedrock Directly
1:59 Strands: The Harness (Agent SDK)
3:14 The Agentic Loop Explained
3:40 Demo: Agent Uses a Tool on Its Own
4:12 AgentCore: The Production Environment
5:24 CLI Walkthrough: Create, Dev, Deploy, Invoke
6:36 AgentCore Extras: Memory, Gateway, Observability
7:26 Recap: Three Layers That Snap Together
#AWS #AmazonBedrock #StrandsAgentsyt:video:S_ZRZFDJ6wES_ZRZFDJ6wEUCT-nPlVzJI-ccQXlxjSvJmwDeploying Agents to the CloudAWS Developershttps://www.youtube.com/channel/UCT-nPlVzJI-ccQXlxjSvJmw2026-07-14T15:29:01+00:002026-07-18T01:39:32+00:00Deploying Agents to the CloudChapter 14 of the Building AI Agent Harnesses with Strands Agents course. Everything up to this point ran on your machine. This chapter is about running it somewhere else, for other people, reliably.
In this video, we'll deploy it to AWS using Amazon Bedrock AgentCore. AgentCore is a collection of components for building, deploying and operating agents in production. The main components we will focus on in this video is AgentCore Runtime and AgentCore Memory.
AgentCore Runtime is where your agent actually lives in production. You containerize your Strands agent, push the image to ECR, and AgentCore Runtime handles everything else: scaling up when requests come in, scaling down when they don't, networking, IAM-based auth, and the long-lived connection management that agent workloads require.
AgentCore Memory is managed persistence for your agent's conversation history and accumulated knowledge across sessions. This is the production-grade version of session managers from Ch 9, without you running your own DynamoDB table, managing TTLs, or designing key schemas.
This course teaches developers how to build AI agents and agent harnesses using the open-source Strands Agents SDK. The course starts by teaching what the difference between an agent and agent harness is, and progresses from fundamental concepts (the agent loop, model providers, tools) through intermediate patterns (hooks, plugins, steering, conversation management) to advanced multi-agent orchestration (agents-as-tools, graph workflows, agent swarms) and finally production operations (guardrails, observability, evaluation, cloud deployment).
Reading: https://go.aws/4eZq6Qz
Github Repo: https://go.aws/4h8siqb
Strands Agents SDK: https://go.aws/3RfeLm9
Follow AWS Developers!
🆇 X: https://go.aws/4plQYhh
💼 LinkedIn: https://go.aws/4vp2edR
This video is a part of a course. Click here to see the whole series: https://go.aws/3SOw7XL
#AWS #AgentHarness #StrandsAgents