<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
        xmlns:news="http://www.google.com/schemas/sitemap-news/0.9">
  <url>
    <loc>https://samtaelabs.com/blog/semantic-code-search-rag-pipeline-parsing-chunking-priority</loc>
    <news:news>
      <news:publication>
        <news:name>삼태연구소</news:name>
        <news:language>ko</news:language>
      </news:publication>
      <news:publication_date>2026-10-05T22:30:03.241Z</news:publication_date>
      <news:title>레거시 코드베이스에 의미론적 검색을 붙이기 전에, 파싱 단계에서 결정해야 할 것들</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://samtaelabs.com/blog/connected-car-data-transparency-consent-design-requirements</loc>
    <news:news>
      <news:publication>
        <news:name>삼태연구소</news:name>
        <news:language>ko</news:language>
      </news:publication>
      <news:publication_date>2026-10-05T12:30:36.229Z</news:publication_date>
      <news:title>커넥티드카 데이터 투명성 설계: 수집 항목 공개부터 제3자 공유 정책까지 요구사항 단계에서 결정해야 할 것들</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://samtaelabs.com/blog/local-llm-tco-vs-cloud-kolibri-comparison</loc>
    <news:news>
      <news:publication>
        <news:name>삼태연구소</news:name>
        <news:language>ko</news:language>
      </news:publication>
      <news:publication_date>2026-10-05T03:00:28.219Z</news:publication_date>
      <news:title>로컬 LLM 도입, 규제 회피가 목적이면 TCO 계산부터 다시 해야 한다: 콜리브리로 보는 비교 기준</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://samtaelabs.com/blog/ai-agent-integrated-workspace-vs-opensource-roi-conditions</loc>
    <news:news>
      <news:publication>
        <news:name>삼태연구소</news:name>
        <news:language>ko</news:language>
      </news:publication>
      <news:publication_date>2026-10-04T22:30:03.243Z</news:publication_date>
      <news:title>AI 에이전트 통합 워크스페이스, 오픈소스 조합보다 나은 조건은 따로 있다</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://samtaelabs.com/blog/ai-agent-documentation-structure-before-memory-plugins</loc>
    <news:news>
      <news:publication>
        <news:name>삼태연구소</news:name>
        <news:language>ko</news:language>
      </news:publication>
      <news:publication_date>2026-10-04T12:30:36.368Z</news:publication_date>
      <news:title>AI 에이전트 도입 전에 메모리 플러그인보다 문서 구조를 먼저 점검해야 하는 이유</news:title>
    </news:news>
  </url>
  <url>
    <loc>https://samtaelabs.com/blog/llm-probabilistic-pattern-agent-design-verification-first</loc>
    <news:news>
      <news:publication>
        <news:name>삼태연구소</news:name>
        <news:language>ko</news:language>
      </news:publication>
      <news:publication_date>2026-10-04T03:00:28.307Z</news:publication_date>
      <news:title>LLM이 &apos;생각한다&apos;는 전제를 버려야 에이전트 설계가 달라진다</news:title>
    </news:news>
  </url>
</urlset>