<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Hypemarc Insights — 에이전트 &amp; AI</title>
    <description>이종 AI 에이전트 오케스트레이션, 클로드코드 노하우, MCP 프로토콜, 사내 도입 거버넌스까지 — Marblo 팀이 정리하는 인사이트</description>
    <link>https://www.hypemarc.com/blog</link>
    <language>ko</language>
    <lastBuildDate>Mon, 25 May 2026 18:01:04 GMT</lastBuildDate>
    <atom:link href="https://www.hypemarc.com/blog/rss.xml" rel="self" type="application/rss+xml"/>
    <item>
      <title>GPT vs Claude vs Gemini — 한국 비즈니스 컨텍스트에서 어떤 모델을 언제 쓸까</title>
      <description>벤치마크는 영어 위주, 사용 사례는 글로벌. 한국어 비즈니스 환경에서 세 모델을 직접 비교한 결과 — 한국어 품질, 비용, 법규 대응, 그리고 *어떤 작업에 어느 모델*.</description>
      <link>https://www.hypemarc.com/blog/gpt-claude-gemini-korean-business</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/gpt-claude-gemini-korean-business</guid>
      <pubDate>Thu, 21 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Claude Code + MCP in Real Workflows — Notes from a Korean AI Studio</title>
      <description>How a Seoul-based AI agency runs Claude Code with MCP servers across every project. The patterns that actually scale, the integrations that paid off, and the workflow tax we eliminated.</description>
      <link>https://www.hypemarc.com/en/blog/claude-code-mcp-korean-ai-studio</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/claude-code-mcp-korean-ai-studio</guid>
      <pubDate>Wed, 20 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>마블로 시작하기 — 첫 멀티 에이전트 만드는 30분 튜토리얼</title>
      <description>Marblo 계정 만들기부터 리서치·작성·검증 3-에이전트 워크플로우를 프로덕션에 배포하기까지 30분. 한국어 환경에서 한 단계씩 따라하는 실전 가이드.</description>
      <link>https://www.hypemarc.com/blog/marblo-first-agent-tutorial</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/marblo-first-agent-tutorial</guid>
      <pubDate>Tue, 19 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Building Your First Multi-Agent System with Marblo — Hands-On Tutorial</title>
      <description>Walk through building a real multi-agent workflow in Marblo, from blank board to production deploy in under thirty minutes. Researcher, writer, fact-checker — heterogeneous models, MCP tools, and observability included.</description>
      <link>https://www.hypemarc.com/en/blog/marblo-first-multi-agent-tutorial</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/marblo-first-multi-agent-tutorial</guid>
      <pubDate>Tue, 19 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>AI 에이전트 도입 전후 3개월 — 우리가 실제로 측정한 숫자</title>
      <description>마케팅·CS·콘텐츠 세 영역에 AI 에이전트를 도입한 3개월. 처리량, 비용, 품질, 만족도 — 자랑이 아니라 실제 측정값을 그대로 공개합니다.</description>
      <link>https://www.hypemarc.com/blog/ai-agent-adoption-3months-data</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/ai-agent-adoption-3months-data</guid>
      <pubDate>Mon, 18 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>MCP Servers in Production — Authentication, Rate Limits, and Observability</title>
      <description>Building MCP (Model Context Protocol) servers for a hobby project is easy. Running them in production with real authentication, real rate limits, and traces you can debug at 2 AM is a different problem. This is what we learned.</description>
      <link>https://www.hypemarc.com/en/blog/mcp-servers-production-guide</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/mcp-servers-production-guide</guid>
      <pubDate>Mon, 18 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Claude Code 실무 도입기 — 4인 팀이 한 달 만에 일하는 방식을 바꾼 기록</title>
      <description>서울의 4인 AI 스튜디오가 Claude Code + MCP 스택을 도입한 첫 한 달. 무엇이 빨라졌고, 어디서 막혔고, 결국 어떤 패턴이 살아남았는가.</description>
      <link>https://www.hypemarc.com/blog/claude-code-team-adoption</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/claude-code-team-adoption</guid>
      <pubDate>Sun, 17 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Heterogeneous Agents in Production — Why Single-Model Setups Fail at Scale</title>
      <description>After running heterogeneous AI agents in production for 18 months, we measured what single-vendor setups give up. The cost premium, the failure modes, and the team-level patterns that only work when you mix models on purpose.</description>
      <link>https://www.hypemarc.com/en/blog/heterogeneous-agents-production</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/heterogeneous-agents-production</guid>
      <pubDate>Sun, 17 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>사내 AI 에이전트 도입 ROI 계산법 — 실전 시트와 함께</title>
      <description>AI 에이전트 도입 비용은 얼마고 언제 회수되는가? 인건비 절감·처리량 증가·오류율 감소를 한 시트에 담는 ROI 계산법을 우리 자체 도입 데이터로 정리합니다.</description>
      <link>https://www.hypemarc.com/blog/ai-agent-roi-calculation</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/ai-agent-roi-calculation</guid>
      <pubDate>Sat, 16 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>AI Agent Orchestration Platforms in 2026 — LangGraph, CrewAI, AutoGen, and Marblo Compared</title>
      <description>An engineering-grade comparison of the major AI agent orchestration platforms in 2026. Where each one shines, where each one breaks, and which choice fits which workload — from prototype to multi-team production.</description>
      <link>https://www.hypemarc.com/en/blog/ai-agent-orchestration-platforms-2026</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/ai-agent-orchestration-platforms-2026</guid>
      <pubDate>Sat, 16 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Codex 5.5 GOAL 모드 분석 — 자율 에이전트의 새로운 표준</title>
      <description>OpenAI Codex의 최신 GOAL 모드가 자율 에이전트의 표준을 어떻게 바꾸고 있는지 분석합니다. 자연어 목표 → 자동 태스크 분해 → 다단계 실행으로 이어지는 패러다임의 의미와, Marblo의 자연어 오케스트레이터와의 비교.</description>
      <link>https://www.hypemarc.com/blog/codex-5-5-goal-mode-analysis</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/codex-5-5-goal-mode-analysis</guid>
      <pubDate>Fri, 15 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Codex 5.5 GOAL Mode — The New Standard for Autonomous Agents</title>
      <description>OpenAI Codex&apos;s latest GOAL mode is reshaping the autonomous agent paradigm. We analyze the shift from imperative commands to goal-driven execution, and how it compares to Marblo&apos;s natural language orchestrator.</description>
      <link>https://www.hypemarc.com/en/blog/codex-5-5-goal-mode-analysis</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/codex-5-5-goal-mode-analysis</guid>
      <pubDate>Fri, 15 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>이종 AI 에이전트가 단일 모델보다 강한 이유 — Claude·GPT·Gemini를 한 보드에서</title>
      <description>왜 2026년 최첨단 AI 팀들은 단일 벤더가 아닌 이종(heterogeneous) 에이전트 오케스트레이션을 선택하는가. Claude의 추론, GPT의 생성, Gemini의 검증 — 모델별 강점 분담이 가져오는 실질적 성능 차이와 비용 효율성.</description>
      <link>https://www.hypemarc.com/blog/heterogeneous-ai-agents-why</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/heterogeneous-ai-agents-why</guid>
      <pubDate>Wed, 13 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Why Heterogeneous AI Agents Beat Single-Model — Claude, GPT, and Gemini on One Board</title>
      <description>Why leading AI teams in 2026 are choosing heterogeneous agent orchestration over single-vendor solutions. The performance gap and cost efficiency that comes from role-based model assignment — Claude reasoning, GPT generation, Gemini verification.</description>
      <link>https://www.hypemarc.com/en/blog/heterogeneous-ai-agents-why</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/heterogeneous-ai-agents-why</guid>
      <pubDate>Wed, 13 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>MCP(Model Context Protocol) 완벽 가이드 — 에이전트가 도구를 갖는 표준</title>
      <description>AI 에이전트가 파일시스템·DB·API·Git 같은 실제 도구에 접근할 수 있게 만든 표준, MCP(Model Context Protocol)를 해부합니다. 왜 이것이 에이전트 산업의 &apos;USB-C&apos;가 됐는지, 사내 시스템 연동 사례까지.</description>
      <link>https://www.hypemarc.com/blog/mcp-protocol-explained</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/mcp-protocol-explained</guid>
      <pubDate>Sun, 10 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Model Context Protocol (MCP) Explained — The Standard for Tool-Wielding Agents</title>
      <description>MCP gives AI agents standardized access to filesystems, databases, APIs, and Git. We unpack why MCP became the &apos;USB-C&apos; of the AI agent industry — and how to integrate it with internal company systems.</description>
      <link>https://www.hypemarc.com/en/blog/mcp-protocol-explained</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/mcp-protocol-explained</guid>
      <pubDate>Sun, 10 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Claude Code 서브에이전트 vs 진짜 멀티 에이전트 — 무엇이 다른가?</title>
      <description>Claude Code의 &apos;subagent&apos; 와 진짜 멀티 에이전트 오케스트레이션의 차이를 해부합니다. 단일 모델 N개 vs 이종 모델 N개, CLI vs 칸반 보드, 메모리 공유 vs 물리적 격리. 어떤 구조가 사내 도입에 적합한가?</description>
      <link>https://www.hypemarc.com/blog/multi-agent-vs-subagent</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/multi-agent-vs-subagent</guid>
      <pubDate>Thu, 07 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Claude Code Subagents vs. Real Multi-Agent Orchestration — What&apos;s the Difference?</title>
      <description>We dissect the gap between Claude Code&apos;s &apos;subagent&apos; pattern and genuine heterogeneous multi-agent orchestration. Single-model N agents vs. heterogeneous N agents, CLI vs. kanban board, shared context vs. physical isolation.</description>
      <link>https://www.hypemarc.com/en/blog/multi-agent-vs-subagent</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/multi-agent-vs-subagent</guid>
      <pubDate>Thu, 07 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>사내 AI 에이전트 운영 거버넌스 5원칙 — 신뢰의 계층을 미리 설계하라</title>
      <description>AI 에이전트를 사내에 도입하면 가장 먼저 부딪히는 질문은 &apos;에이전트가 작성한 코드를 누가 책임지는가&apos; 입니다. 사내 운영을 위한 5가지 거버넌스 원칙: 권한 분리, 감사 로그, 롤백 경로, 측정 지표, 팀원 교육.</description>
      <link>https://www.hypemarc.com/blog/ai-agent-governance-5-principles</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/ai-agent-governance-5-principles</guid>
      <pubDate>Mon, 04 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>5 Principles for In-house AI Agent Governance — Design the Trust Hierarchy First</title>
      <description>When you adopt AI agents internally, the first question is &apos;Who is responsible for code an agent wrote?&apos; This article lays out 5 governance principles: privilege separation, audit logs, rollback paths, measurable KPIs, and team training.</description>
      <link>https://www.hypemarc.com/en/blog/ai-agent-governance-5-principles</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/ai-agent-governance-5-principles</guid>
      <pubDate>Mon, 04 May 2026 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>업종별 AI 에이전트 구축 사례 8선: 이커머스, SaaS, 금융, 제조업까지</title>
      <description>8가지 업종의 실제 AI 에이전트 구축 사례를 소개합니다. 각 업종별로 어떤 에이전트를 어떻게 구축했는지, 구체적인 아키텍처와 성과 데이터를 공개합니다.</description>
      <link>https://www.hypemarc.com/blog/ai-agent-use-cases-by-industry</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/ai-agent-use-cases-by-industry</guid>
      <pubDate>Tue, 25 Feb 2025 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>2025 AI 에이전트 구축 방법론: OpenAI Agent SDK부터 멀티에이전트 아키텍처까지</title>
      <description>최신 AI 에이전트 구축 기술을 총정리합니다. OpenAI Agent SDK, Claude MCP, Function Calling, RAG 파이프라인, 멀티에이전트 오케스트레이션 등 실전에서 사용하는 아키텍처와 구축 방법론을 다룹니다.</description>
      <link>https://www.hypemarc.com/blog/ai-agent-architecture-2025</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/ai-agent-architecture-2025</guid>
      <pubDate>Sun, 23 Feb 2025 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>Building AI Agents for the Korean Market: Platforms, Regulations, and Real-World Use Cases</title>
      <description>How to build and deploy AI agents optimized for Korean business environments. Covers KakaoTalk integration, Korean NLP challenges, compliance requirements, and successful implementation strategies for international companies entering Korea.</description>
      <link>https://www.hypemarc.com/en/blog/ai-agent-building-korean-market</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/ai-agent-building-korean-market</guid>
      <pubDate>Wed, 12 Feb 2025 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>N8N vs Flask: 왜 우리는 파이썬 코드 기반으로 AI 에이전트를 구축하는가</title>
      <description>N8N, Make 같은 노코드 자동화 플랫폼과 Flask+Python 기반 에이전트 구축의 차이를 비교합니다. 실제 프로젝트 경험을 바탕으로 코드 기반 접근이 기업용 에이전트에 더 적합한 이유를 설명합니다.</description>
      <link>https://www.hypemarc.com/blog/flask-python-ai-agent-building</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/flask-python-ai-agent-building</guid>
      <pubDate>Mon, 10 Feb 2025 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>AI 챗봇 vs 전통 CS: 비용 절감과 고객 만족도 비교 분석</title>
      <description>AI 챗봇과 전통적인 고객 서비스를 비용, 응답 속도, 고객 만족도, 확장성 4가지 축으로 비교 분석합니다. 하이브리드 CS 전략으로 최적의 고객 경험을 설계하세요.</description>
      <link>https://www.hypemarc.com/blog/ai-chatbot-vs-traditional-cs</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/ai-chatbot-vs-traditional-cs</guid>
      <pubDate>Thu, 06 Feb 2025 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>AI-Powered Customer Service: Complete ROI Analysis for 2025</title>
      <description>A detailed ROI analysis comparing AI-powered customer service with traditional support teams. Includes cost breakdowns, satisfaction metrics, implementation roadmap, and real-world case studies.</description>
      <link>https://www.hypemarc.com/en/blog/ai-customer-service-roi</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/ai-customer-service-roi</guid>
      <pubDate>Sat, 01 Feb 2025 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>AI 에이전트 도입 성공 사례: 법무법인부터 이커머스까지</title>
      <description>다양한 산업에서 AI 에이전트를 도입해 실질적인 비즈니스 성과를 낸 사례를 분석합니다. 법무법인, 이커머스, 금융, 교육 등 업종별 구체적인 ROI와 도입 과정을 확인하세요.</description>
      <link>https://www.hypemarc.com/blog/ai-agent-implementation-cases</link>
      <guid isPermaLink="true">https://www.hypemarc.com/blog/ai-agent-implementation-cases</guid>
      <pubDate>Tue, 28 Jan 2025 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
    <item>
      <title>AI Agents for Business Automation: Complete Guide 2025</title>
      <description>Learn how to automate customer service, sales, and marketing with AI agents using OpenAI&apos;s Agent Builder, Response API, and custom implementations.</description>
      <link>https://www.hypemarc.com/en/blog/ai-agents-business-automation</link>
      <guid isPermaLink="true">https://www.hypemarc.com/en/blog/ai-agents-business-automation</guid>
      <pubDate>Fri, 10 Jan 2025 00:00:00 GMT</pubDate>
      <category>AI Agents</category>
    </item>
  </channel>
</rss>