#AWS
43 related articles
Lazy AWS: A Keyboard-Driven Terminal for AWS Resource Management
Lazy AWS is an open-source, keyboard-driven terminal tool that lets developers browse and inspect AWS resources and run confirmed AWS CLI operations without leaving the terminal.
DynamoDB in Practice: From Creating a Table via CLI to Understanding the Cost of Every Read
Hands-on with DynamoDB via AWS CLI: create a key-value table, write typed JSON, compare GetItem/Query/Scan capacity, and see how the partition key drives read costs.
AgentCore Memory in Practice: How Short-Term and Long-Term Memory Work Across Sessions
A deep dive into AWS AgentCore's short- and long-term memory: USER_PREFERENCE, EPISODIC, and SEMANTIC strategies for cross-session preference retrieval and actor-level isolation, with measured assertions, extraction latency, and cost data.
Farewell to Long-Term Keys: Passwordless AWS CLI Login and Policy Simulator in Practice
Log in to the AWS CLI without long-term access keys: use aws login for auto-refreshing temporary credentials, then verify policy allow and deny behavior with the IAM policy simulator—no real requests required. Includes version requirements, command examples, result interpretation, and cleanup.
AWS Lambda Web Functions Tested: Can It Replace Amplify for Next.js Deployments?
AWS Lambda Web Functions tested against Amplify: TTFB, checkout, and order page performance data for the same Next.js app on both hosting options, analyzing streaming, concurrency models, and CDN trade-offs.
Amazon S3 Officially Launches on the Vercel Marketplace
Amazon S3 joins the Vercel Marketplace as an official integration, enabling S3 resource creation and connection within Vercel, using OIDC short-lived credentials instead of static AWS keys, with v0 code generation support.
Getting Started with SageMaker AI Notebooks: Building Agents with Strands Agents
Build an AI Agent from scratch on an AWS SageMaker AI notebook with the open-source Strands Agents SDK, calling a calculator and weather API. Covers setup, tool definition, Bedrock permissions, and cost control.
AWS regional pricing has no uniform multiplier: model price gaps within the same region can reach 1.6x
A full audit of 1,586 Bedrock billing items shows 32 regions produce 123 distinct multipliers—there is no uniform regional markup. Within one region, multipliers range from 0.964 to 1.700; London alone has 25.
JalOS Deep Dive: Building a Predictive Water Management Platform with FastAPI, Flutter, React, and AWS
JalOS is an AI-powered predictive water management platform built with FastAPI, Flutter, React, and AWS. Explore its three-tier architecture, Ridge Regression forecasting, water duration estimation, and Terraform + GitHub Actions OIDC deployment.
AWS Generative AI Landscape: Tools and Infrastructure Breakdown
A breakdown of AWS tools for generative AI — covering Bedrock, SageMaker, and EC2 accelerator chips — with a full workflow from data prep to deployment and cost optimization tips.
AWS AgentCore Gateway in Practice: Extending AI Agent Capabilities with MCP
A hands-on guide to AWS AgentCore Gateway (MCP Gateway): covers MCP protocol basics, Target Types, Signature V4 auth, OAuth config, and Secrets Manager credential management.
AWS September Updates: Bedrock and AgentCore Bring New AI Development Capabilities
AWS September updates to Amazon Bedrock, AgentCore, and Strands bring broader model selection, faster serverless agents with built-in evaluation, and automatic knowledge base sync with native enterprise connectors.
SageMaker HyperPod: Isolation and Fairness in Cross-Team GPU Cluster Sharing
Amazon SageMaker HyperPod's reference architecture for cross-team GPU cluster sharing uses IAM Identity Center, Kubernetes namespace isolation, Task Governance, and cost chargeback to enable secure, fair, and cost-transparent multi-tenant compute.
Claude Haiku 5.5 Arrives on AWS: A High-Efficiency Model with ~75% Lower Costs
Claude Haiku 5.5 is now on Amazon Bedrock and AWS. Anthropic's fastest, most efficient Claude 5.5 model cuts costs ~75% vs Haiku 4.5, optimized for subagents and high-volume workloads.
AWS DevOps Agent Auto-Remediation in Practice: From Diagnosis to One-Click Approval
AWS DevOps Agent only diagnoses by default. Learn how Lambda Durable Functions, EventBridge, and Bedrock turn investigation summaries into pre-validated fixes for one-click approval.
PyTorch Performance Profiling in Practice: From torch.profiler to AWS Trainium Hardware Cycles
AWS Annapurna Labs engineers share a practical PyTorch profiling methodology: capturing traces with torch.profiler on AWS Trainium, drilling into hardware cycles, tracing dependency chains, and using automated analysis to identify host overhead, communication stalls, and idle compute.
SageMaker Studio Now Supports HyperPod Spaces Management — A Better Developer Experience
Amazon SageMaker Studio now supports managing HyperPod EKS Spaces directly in the UI — create, start, stop, and open JupyterLab and Code Editor environments without the command line.
Vibe Coding Demo Day: Pitching AI-Built Apps for Virtual Millions
At an AWS-sponsored vibe coding event, developers pitched AI-built apps in a mock Shark Tank format. See what sold, who overspent, and what it reveals about AI-assisted development.
AWS MCP Server Expands to Six New Regions: Shorter Paths for Agent Calls
AWS MCP Server now covers 8 execution regions across the US, Asia Pacific, and Europe, enabling AI agents to connect locally, reduce latency, and meet data residency requirements.
Downgrading User Roles in Amazon QuickSight: Two Reliable Methods Explained
Amazon QuickSight offers no direct console path to downgrade user roles. This guide explains two reliable methods: manual delete-and-recreate, and a safe AWS CLI in-place downgrade that preserves asset ownership.