O que eu publico no AWS Builder Center, os badges que a comunidade dá por isso e o que outros builders estão escrevendo sobre agentes de IA. Os artigos daqui vão para lá; os bons de lá aparecem aqui.
AWS Builder Center
03 de out. de 2026
On October 1, 2026, Security Hub gained remediation plans: instead of a queue of findings, one plan per root cause, with an impact assessment and instructions in CLI, Terraform, CloudFormation, Python, and CDK. I dissect the root-cause remediation pattern (the problem it solves, its anatomy, when to use it and when not) and the detail that deserves the most scrutiny: AI agents consuming plans through the API.
03 de out. de 2026
AgentCore Harness reached GA in June 2026 as a managed abstraction that collapses the LLM agent control plane into two API calls. In this article, I analyze how the harness works internally, where it fails, and what architects of financial-grade systems need to understand before putting it into production.
Badges
Knowledge Seeker
Read 10 different articles.
out. de 2026
Discussion Debut
Join the conversation by commenting on content.
out. de 2026
First Article
Publish your first article.
out. de 2026
Photo Finisher
Upload a photo to your profile.
mar. de 2026
Hello, World!
Complete the About section in your profile to introduce yourself in Builder Center.
mar. de 2026
Da comunidade
Atualizado a cada 6 horas, só de AWS Heroes, Community Builders e gente da AWS. O texto continua no Builder Center, com os autores.
Hazel SáenzAWSAcabas de construir un agente genial. Corre en tu máquina, le preguntas algo y te responde justo como querías. El siguiente paso natural es que otras personas lo usen: enseñárselo a tus amigos, ponerlo en línea, que viva en algún lado que no sea tu terminal.
Mustkhim InamdarCommunity BuilderThe real cost math, the guardrails that stopped it breaking prod, and the failure modes nobody documents. A field report on AWS DevOps Agent vs rolling your own on Bedrock AgentCore.
Vijayaraghavan VashudevanCommunity BuilderThe one rule that holds everything together: Never trust the prompt. Never blindly trust the model output. Never give an AI agent more access than it needs.
Ajaykumar K VCommunity BuilderBuild a production-ready AI Learning Assistant on Amazon Bedrock, one layer at a time. This first part introduces the architecture, explains why AgentCore was chosen over legacy Bedrock Agents, and sets up Amazon Nova with intelligent model routing. It also covers the project scaffold and a preflight “doctor” that validates AWS credentials, region, model access, and end-to-end inference before development begins.
asahideCommunity BuilderBy adding IAM conditions that apply only to calls through the AWS MCP Server, you can let an AI agent work with your own credentials while narrowing what it can do to what the task needs. I tested this with two agent tasks: a support investigation and a sales report.
In this post, we walk through building a governed data agent by integrating Databricks Genie with Amazon Bedrock AgentCore, covering steps from deploying a gateway and seeding a Genie Agent with data, to hosting on AgentCore Runtime, managing secrets, and validating governance, enabling secure, production-ready AI agents that query Databricks data with proper access controls.
Escreveu sobre agentes de IA, Bedrock ou arquitetura na AWS? Me siga no Builder Center ou me marque no post: eu leio e comento com o que vi em produção. Se você é Community Builder, Hero ou da AWS e usa a tag #ai-agents ou #amazon-bedrock-agentcore, o post já aparece na seção acima.
