AI Briefing
AWS Tech Blog (Korea) news
The latest AI and developer news from AWS Tech Blog (Korea), with short summaries.
Visit website →How Karrot Built a Large-Scale Signing Key Management System with AWS CloudHSM – Part 2: CloudHSM Architecture and Multi-Layer Access Control
Karrot leveraged AWS CloudHSM to design a security architecture with three layers of access control across the network, application, and HSM internals.
How Danggeun Built a Large-Scale Signing Key Management System with AWS CloudHSM – Part 1: The Importance of Signing Key Security and Service Selection
Danggeun introduced a hybrid system combining AWS CloudHSM and KMS to securely process an average of 65 million JWT signings per day.
Eartth Alliance's CS Refund Auto-Response Agent System Built in 2 Weeks with Strands SDK and Amazon Bedrock AgentCore
Eartth Alliance built a CS refund auto-response system based on Amazon Bedrock AgentCore in 2 weeks through the AWS DEVCRAFT hackathon.
Beyond Personal Productivity to Organizational Productivity: A 10-Video Learning Plan for Claude Code on Amazon Bedrock
This introduces a learning plan based on Amazon Bedrock for organizational management and control, going beyond individual-level use of Claude Code.
Building an Amazon Connect Customer AI Agent with a 1-Hour Conversation [AICC Builder Series #1]
AWS has released AICC Builder, an open-source tool that lets you build a PoC for an Amazon Connect-based AI phone agent with no code, using only a 1-hour interview.
Optimizing EKS Costs with AWS Graviton
Adopting ARM64-based AWS Graviton instances in EKS can cut costs by up to 40% compared to x86.
A Detailed Look at Claude Apps Gateway on AWS
Anthropic's LLM Gateway built into Claude Code was deployed directly in an AWS environment, with its features and limitations examined in detail.
Financial Cloud Guide A to Z Part 2 – R&D Network Exceptions and the Network Separation Improvement Roadmap
Following the 2024 Network Separation Improvement Roadmap, this guide walks financial companies step by step through safely implementing R&D networks using an AWS multi-account architecture.
Financial Cloud Guide A to Z Part 1 – First Steps in Cloud Adoption Explained Through the Electronic Financial Supervisory Regulation
This piece organizes, from a practical perspective, the 7-step process and safety assurance measures for cloud adoption by financial companies based on the Electronic Financial Supervisory Regulation.
AWS Transform for migrations now supports localization in 13 languages
The AWS Transform migration workflow has started supporting localization in 13 languages, including Korean.
AWS Compute Selection Guide for Distributed Training (Part 1: Model Scale and Hardware Selection)
This covers how to select AWS compute resources optimized for model scale and hardware specifications for LLM training.
Kiro Agent and Harness Engineering: Automating Large-Scale Knowledge-Based Migration at LGU+
LGU+ leveraged Kiro agent and Harness Engineering to streamline AWS cloud migration for over 100 services.
Building an In-House LLM Gateway on Amazon Bedrock: Authentication, Cost Management, and Governance for Claude Code and Codex
In preparation for the spread of AI tools like Claude Code, this article proposes a way to build an LLM Gateway centered on authentication, cost control, and governance using Amazon Bedrock.
Validating Production-Ready Diffusion Language Models (DLM) on AWS
It covers evaluation gate design to verify response quality and RAG utilization for production adoption of DLMs with low latency.
Amazon CloudWatch Introduces OpenTelemetry and PromQL Support
Amazon CloudWatch has launched general availability of OpenTelemetry metrics ingestion and PromQL query support, strengthening observability.
PostgreSQL 18 on Amazon Aurora and Amazon RDS: Performance Improvements
PostgreSQL 18 significantly improves query performance and management efficiency through skip scan optimization and enhanced EXPLAIN features.
StradVision's AWS Cloud-Based Physical AI End-to-End Pipeline Acceleration Case
StradVision built a synthetic data generation pipeline for rare scenarios by leveraging AWS cloud and a hybrid architecture.
Evaluating Amazon Bedrock Model Performance with promptfoo
This introduces a method for systematically evaluating the response quality of Amazon Bedrock models using the open-source framework promptfoo.
Oracle Database@AWS Network Configuration Guide
This explains the key characteristics and configuration options for designing an efficient network architecture when adopting Oracle Database@AWS.
The Rules of Uninterrupted AWS Site-to-Site VPN Operations: Preventing Maintenance Outages with BGP and Lifecycle Control
When operating AWS Site-to-Site VPN, you can use BGP and Lifecycle Control to prevent communication outages caused by maintenance.