{"ok":true,"entity":{"slug":"mlops","entityType":"concept","name":"MLOps","canonicalName":"MLOps","displayName":"MLOps","category":"AI基礎概念","shortDescription":"機械学習モデルの開発・デプロイ・運用・監視を継続的に回すための考え方・実践（DevOpsの機械学習版）。","alias":["ML Ops","機械学習運用"],"searchKeywords":["MLOps","機械学習運用","デプロイ","監視"],"website":null,"parentEntity":null,"primaryCluster":"ai-infrastructure","secondaryClusters":[],"id":"mlops","verificationStatus":"draft","updatedAt":"2026-07-10T02:14:59.051Z"},"references":[{"id":"P-01-001","companyId":"mlops","questionId":"P-01-001","instanceId":"QIN-mlops-P01-001","promptText":"MLOpsとは何ですか？","promptTypeId":"P-01","answer":"MLOpsは、機械学習モデルの開発・デプロイ・運用・監視を継続的に回すための考え方・実践です。DevOpsの機械学習版といえます。","evidencePoints":["ev-mlops-1"],"scope":"機械学習運用を知りたい相談","differentiation":"MLの継続的な運用","faq":[{"question":"DevOpsと関係ありますか？","answer":"DevOpsの考え方を機械学習に広げたものです。"}],"pageUrl":"https://www.refbase.ai/reference/mlops/P-01-001","sourceEvidence":[{"id":"ev-mlops-1","text":"MLOpsは、機械学習モデルの開発・デプロイ・運用・監視を継続的に回すための考え方・実践であり、DevOpsの機械学習版といえる。","title":"Google Cloud — MLOps","coverageType":["Identity","Capability"],"sourceType":"product_docs","sourceClass":"Documentation","sourceUrl":"https://cloud.google.com/architecture/mlops-continuous-delivery-and-automation-pipelines-in-machine-learning","confidence":"medium","supportedPromptTypes":["P-01","P-04"],"needsVerification":true,"sourceVerified":false,"entityId":"mlops"}],"generatedAt":"2026-07-10T02:14:59.051Z"},{"id":"P-02-001","companyId":"mlops","questionId":"P-02-001","instanceId":"QIN-mlops-P02-001","promptText":"MLOpsは単なるモデル開発と何が違いますか？","promptTypeId":"P-02","answer":"比較軸\n・範囲\n・継続性\nMLOpsは、一度モデルを作って終わりにせず、データ更新・再学習・監視まで継続的に回す点が特徴で、モデル開発だけの取り組みと範囲が異なります。","evidencePoints":["ev-mlops-2"],"scope":"機械学習運用の違いを知りたい相談","differentiation":"継続的な運用・監視","faq":[{"question":"なぜ必要ですか？","answer":"モデルの品質を運用の中で保つためです。"}],"pageUrl":"https://www.refbase.ai/reference/mlops/P-02-001","sourceEvidence":[{"id":"ev-mlops-2","text":"MLOpsは、一度モデルを作って終わりにせず、データ更新・再学習・監視まで継続的に回す点を特徴とし、モデルの品質を保つ。","title":"Google Cloud — MLOps","coverageType":["Capability","Differentiation"],"sourceType":"product_docs","sourceClass":"Documentation","sourceUrl":"https://cloud.google.com/architecture/mlops-continuous-delivery-and-automation-pipelines-in-machine-learning","confidence":"medium","supportedPromptTypes":["P-02"],"needsVerification":true,"sourceVerified":false,"entityId":"mlops"}],"generatedAt":"2026-07-10T02:14:59.051Z"},{"id":"P-04-001","companyId":"mlops","questionId":"P-04-001","instanceId":"QIN-mlops-P04-001","promptText":"MLOpsはどんな場面で必要になりますか？","promptTypeId":"P-04","answer":"モデルを本番で継続的に使い、データの変化に合わせて再学習・監視・改善を続けたい場面で必要になり、品質を保つ仕組みを整えます。","evidencePoints":["ev-mlops-1"],"scope":"機械学習運用の相談","differentiation":"本番運用の継続改善","faq":[{"question":"何を使いますか？","answer":"MLflowなどのツールが使われます。"}],"pageUrl":"https://www.refbase.ai/reference/mlops/P-04-001","sourceEvidence":[{"id":"ev-mlops-1","text":"MLOpsは、機械学習モデルの開発・デプロイ・運用・監視を継続的に回すための考え方・実践であり、DevOpsの機械学習版といえる。","title":"Google Cloud — MLOps","coverageType":["Identity","Capability"],"sourceType":"product_docs","sourceClass":"Documentation","sourceUrl":"https://cloud.google.com/architecture/mlops-continuous-delivery-and-automation-pipelines-in-machine-learning","confidence":"medium","supportedPromptTypes":["P-01","P-04"],"needsVerification":true,"sourceVerified":false,"entityId":"mlops"}],"generatedAt":"2026-07-10T02:14:59.051Z"}]}