[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"portal-settings:stajic:en":3,"public-menus:all":38,"post:ai-agent-reliability-why-the-final-answer-is-not-enough:en":205,"related:post:ai-agent-reliability-why-the-final-answer-is-not-enough:en:1":791},{"statusCode":4,"data":5,"message":37},200,{"tenantId":6,"lang":7,"defaultLang":8,"siteUrl":9,"contactEmail":10,"brandName":11,"logoUrl":12,"siteName":11,"siteDescription":13,"ogImage":10,"robotsIndex":14,"socialLinks":10,"reservedSlugs":10,"seoPolicy":15},"stajic","en","de","https:\u002F\u002Fstajic.de",null,"Stajic Platform","\u002FLogo_Planet.svg","Stajic Portal",true,{"branding":16,"relatedContent":17,"crossDomainLinks":18},{"logoUrl":12},{"enabled":14},[19,22,25,28,31,34],{"url":20,"label":21,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Ffigure.rocks","figure.rocks",{"url":23,"label":24,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Floving.rocks","loving.rocks",{"url":26,"label":27,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.com","bazify.com",{"url":29,"label":30,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.de","bazify.de",{"url":32,"label":33,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.at","bazify.at",{"url":35,"label":36,"isActive":14,"showInFooter":14,"includeInSameAs":14},"https:\u002F\u002Fbazify.ba","bazify.ba","Portal settings resolved",[39,45],{"id":40,"name":41,"location":42,"isActive":14,"isDefault":43,"items":44},1,"main-navigation","header",false,[],{"id":46,"name":47,"location":48,"isActive":14,"isDefault":14,"items":49},4,"main-menu","sidebar",[50,66,79,93,103,118,133],{"id":51,"title":52,"url":60,"target":61,"icon":62,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":64,"portfolioId":10,"children":65},"item-18",{"de":53,"en":54,"es":55,"fr":56,"it":54,"ru":57,"sr":58,"zh":59},"Startseite","Home","Inicio","Accueil","Главная","Почетна","首页","\u002Ffull-stack-web-developer-munich-performance-seo-and-maintainable-builds","_self","i-lucide-home","page",111,[],{"id":67,"title":68,"url":75,"target":61,"icon":76,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":77,"portfolioId":10,"children":78},"item-22",{"de":69,"en":69,"es":70,"fr":69,"it":71,"ru":72,"sr":73,"zh":74},"Vision","Visión","Visione","Видение","Визија","想象","\u002Fueber-uns-webdesign-muenchen-webaplikation","i-lucide-eye",113,[],{"id":80,"title":81,"url":89,"target":61,"icon":90,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":91,"portfolioId":10,"children":92},"item-19",{"de":82,"en":83,"es":84,"fr":83,"it":85,"ru":86,"sr":87,"zh":88},"Leistungen","Services","Servicios","Servizi","Услуги","Услуге","服务","\u002Fservices-dienstleistungen-muenchen","i-lucide-wrench",116,[],{"id":94,"title":95,"url":99,"target":61,"icon":100,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":101,"portfolioId":10,"children":102},"item-23",{"de":96,"en":96,"es":96,"fr":96,"it":96,"ru":97,"sr":97,"zh":98},"Blog","Блог","博客","\u002Fblog","i-lucide-book-open",112,[],{"id":104,"title":105,"url":114,"target":61,"icon":115,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":116,"portfolioId":10,"children":117},"item-32",{"de":106,"en":107,"es":108,"fr":109,"it":110,"ru":111,"sr":112,"zh":113},"Neue Technologien","New Technologies","Nuevas tecnologías","Nouvelles technologies","Nuove tecnologie","Новые технологии","Нове технологије","新技术！","\u002Fneue-webtechnologien","i-lucide-sparkles",122,[],{"id":119,"title":120,"url":129,"target":61,"icon":130,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":131,"portfolioId":10,"children":132},"item-20",{"de":121,"en":122,"es":123,"fr":124,"it":125,"ru":126,"sr":127,"zh":128},"Kontakt","Contact us!","Contacto","Contact","Contatto","Контакт","Контактирајте нас","联系我们！","\u002Fcontact","i-lucide-mail",115,[],{"id":134,"title":135,"url":144,"target":61,"icon":145,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":146,"portfolioId":10,"children":147},"item-21",{"de":136,"en":137,"es":138,"fr":139,"it":140,"ru":141,"sr":142,"zh":143},"Unsere Arbeit","Our Work","Nuestro trabajo","Nos réalisations","I nostri lavori","Наши работы","Наши радови","文件夹","\u002Fportfolio","i-lucide-briefcase",114,[148,161,175,181,193],{"id":149,"title":150,"url":144,"target":61,"icon":159,"isActive":14,"type":63,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":146,"portfolioId":10,"children":160},"item-24",{"de":151,"en":152,"es":153,"fr":154,"it":155,"ru":156,"sr":157,"zh":158},"Alle Projekte","All Projects","Todos los proyectos","Tous les projets","Tutti i progetti","Все проекты","Сви пројекти","所有项目","i-lucide-grid-3x3",[],{"id":162,"title":163,"url":171,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":174},"item-29",{"de":164,"en":165,"es":166,"fr":167,"it":168,"ru":169,"sr":170,"zh":143},"Local Roots, Global Reach","Local Roots - Global Reach","Empresa local ","Entreprise locale","Azienda locale","Местная компания","Локално предузеће глобално тржиште","\u002Fportfolio\u002Flocal-roots-global-reach-communication-media-systems-for-modern-business","i-lucide-folder","custom",[],{"id":176,"title":177,"url":179,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":180},"item-28",{"de":178,"en":178,"es":178,"fr":178,"it":178,"ru":178,"sr":178,"zh":178},"Solr Suggester","\u002Fportfolio\u002Fsolr-fuzzy-suggester-und-solr-infix-suggester-abfrage-ueber-ajax-und-filterung",[],{"id":182,"title":183,"url":191,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":192},"item-27",{"de":184,"en":185,"es":186,"fr":187,"it":188,"ru":189,"sr":190,"zh":185},"Firmenwebseite SEO","Company Website SEO","Sitio web corporativo SEO","Site web d’entreprise SEO","Sito web aziendale SEO","Корпоративный сайт SEO","Пословна веб-страница SEO","\u002Fportfolio\u002Fseo-sem-branding-mobile-webseite-muenchen",[],{"id":194,"title":195,"url":203,"target":61,"icon":172,"isActive":14,"type":173,"productId":10,"categoryId":10,"shopCategoryId":10,"articleId":10,"pageId":10,"portfolioId":10,"children":204},"item-31",{"de":196,"en":197,"es":198,"fr":199,"it":200,"ru":201,"sr":202,"zh":197},"Digitalisierungsportal","Digitalization Portal","Portal de digitalización","Portail de numérisation","Portale di digitalizzazione","Портал цифровизации","Портал за дигитализацију","\u002Fportfolio\u002Fdigitalisierungsportal-archiv-museum-bibliothek-ead-lido-mets-mods",[],{"statusCode":4,"data":206,"message":790},{"id":207,"title":208,"slug":209,"content":210,"contentJson":211,"excerpt":561,"featuredImage":562,"featuredImageAlt":563,"featuredImageCaption":10,"featuredImageTitle":10,"featuredImageCopyright":10,"featuredImageAuthor":10,"featuredImageSourceUrl":10,"featuredImageLicense":10,"featuredImageIsAiGenerated":43,"status":564,"publishedAt":565,"createdAt":566,"updatedAt":567,"seoLocalePaths":568,"categories":577,"author":586,"translations":591},"460","AI Agent Reliability: Why the Final Answer Is Not Enough","ai-agent-reliability-why-the-final-answer-is-not-enough","{\"time\":1788955485785,\"blocks\":[{\"data\":{\"text\":\"\u003Cb>Correct output does not prove correct reasoning, safe execution, or a trustworthy system.\u003C\u002Fb>\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"For years, AI evaluation has been dominated by a deceptively simple question: \u003Cb>Was the answer correct?\u003C\u002Fb> For a chatbot, this may sometimes be sufficient. For an agent capable of searching systems, reading data, calling tools, modifying state, executing workflows, writing files, interacting with APIs, or making decisions, it is not.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"An agent can produce the correct final answer while doing several things wrong on the way there. It can use the wrong source, misunderstand an instruction and later compensate for the mistake, access unnecessary information, execute an unauthorized intermediate action, silently recover from an error that should have triggered escalation, or leave behind side effects nobody noticed.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"That creates one of the central problems of agentic AI: \u003Cb>a correct outcome does not prove a correct trajectory.\u003C\u002Fb>\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"The Outcome Illusion\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Traditional software gives us an intuitive model of correctness. Input enters a deterministic or mostly deterministic system, logic is executed, output is produced, and tests verify expected behavior. LLM-based systems weaken this assumption. Agentic systems go further.\"},\"type\":\"paragraph\"},{\"data\":{\"items\":[\"model interpretation\",\"retrieved context\",\"tool selection\",\"intermediate observations\",\"external state\",\"previous actions\",\"model-generated plans\",\"permission boundaries\",\"retries and fallback behavior\",\"human interaction\"],\"style\":\"unordered\"},\"type\":\"list\"},{\"data\":{\"text\":\"Two executions starting from nearly identical inputs may reach the same result through very different paths. If evaluation observes only the final output, most of the system remains invisible.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Imagine an AI agent receives the instruction: \u003Ci>Update the customer's billing address.\u003C\u002Fi> The address is ultimately updated correctly. A conventional evaluation might classify the task as successful.\"},\"type\":\"paragraph\"},{\"data\":{\"items\":[\"The agent searches several unrelated customer records.\",\"It retrieves more personal information than required.\",\"It initially modifies the wrong account.\",\"It notices the mistake.\",\"It reverses the change.\",\"It updates the correct account.\",\"It reports success.\"],\"style\":\"ordered\"},\"type\":\"list\"},{\"data\":{\"text\":\"\u003Cb>Final state: correct. System behavior: unacceptable.\u003C\u002Fb> An outcome-only benchmark gives this execution a pass. A production assurance system should not.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"The Trajectory Is Part of the Product\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"This is why the \u003Cb>trajectory\u003C\u002Fb> of an AI agent must become a first-class engineering object. A trajectory is the sequence of relevant states and actions between the original request and the final result.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"Intent → Context → Decision → Tool → Action → Observation → Decision → State change → Result\"},\"type\":\"code\"},{\"data\":{\"text\":\"Zachary J. Stevens develops this idea in \u003Ci>The Trajectory Is the System\u003C\u002Fi>, arguing that agentic evaluation must move beyond the final answer and examine the complete path of action through a changing environment.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"A correct outcome does not excuse an unacceptable trajectory.\",\"caption\":\"Zachary J. Stevens, The Trajectory Is the System\",\"alignment\":\"left\"},\"type\":\"quote\"},{\"data\":{\"link\":\"https:\u002F\u002Fzacharyjstevens.com\u002Fdispatches\u002Fvanguard-signal\u002F009-the-trajectory-is-the-system\u002F\",\"meta\":{\"image\":{},\"title\":\"The Trajectory Is the System\",\"description\":\"Zachary J. Stevens — DFEI.009 on evaluating agentic systems by their complete trajectory rather than only the final outcome.\"}},\"type\":\"linkTool\"},{\"data\":{\"text\":\"The distinction matters enormously. Reliability is therefore not simply \u003Cb>correct output\u003C\u002Fb>. It is closer to \u003Cb>acceptable outcome + acceptable trajectory + recoverability + evidence\u003C\u002Fb>.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"A Correct Answer Can Hide a Broken System\",\"level\":2},\"type\":\"header\"},{\"data\":{\"content\":[[\"Agent\",\"Final result\",\"Execution\"],[\"A\",\"Correct\",\"Correct path\"],[\"B\",\"Correct\",\"Unsafe path\"],[\"C\",\"Incorrect\",\"Safe failure\"],[\"D\",\"Incorrect\",\"Unsafe failure\"]],\"withHeadings\":true},\"type\":\"table\"},{\"data\":{\"text\":\"Most benchmark-driven evaluation strongly rewards A and B and penalizes C and D. Operationally, however, \u003Cb>B can be more dangerous than C\u003C\u002Fb>. Agent C may recognize uncertainty, stop execution and request human review. Agent B may confidently produce correct results while violating assumptions that nobody is monitoring.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"successful output → increased trust → broader permissions → more automation → larger blast radius\"},\"type\":\"code\"},{\"data\":{\"text\":\"We Need Evidence, Not Confidence\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"One of the biggest mistakes in AI adoption is treating model confidence, user satisfaction or historical success rate as evidence of system reliability. They are not equivalent.\"},\"type\":\"paragraph\"},{\"data\":{\"items\":[\"What did the agent receive?\",\"What context did it retrieve?\",\"Which tools did it call?\",\"Why was the action allowed?\",\"What state existed before the action?\",\"What changed?\",\"Which intermediate failures occurred?\",\"Was anything retried?\",\"Was human approval required?\",\"Could execution have been stopped?\",\"Can the action be reversed?\",\"Which model, prompt and tool versions were involved?\"],\"style\":\"unordered\"},\"type\":\"list\"},{\"data\":{\"text\":\"Without these answers, there is no serious operational assurance. There is only an output. Observability and evidence must therefore be designed into agent architecture rather than added after deployment.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Logging Is Not the Same as Control\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Organizations often respond: \u003Ci>Everything is logged.\u003C\u002Fi> Good. But logging alone does not control anything. A log tells you what happened. A control determines whether something \u003Cb>may happen\u003C\u002Fb>.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"Agent requests DELETE \u002Fcustomer\u002F123 ↓\\nAction logged ↓\\nDELETE executed\"},\"type\":\"code\"},{\"data\":{\"text\":\"That gives observability. Compare it with:\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"Agent requests DELETE \u002Fcustomer\u002F123 ↓\\nPolicy evaluation ↓\\nCurrent identity verified ↓\\nCurrent action parameters checked ↓\\nRisk threshold evaluated ↓\\nHuman approval if required ↓\\nAction executed ↓\\nResult verified ↓\\nEvidence stored\"},\"type\":\"code\"},{\"data\":{\"text\":\"Now we are approaching a control system. The difference is architectural, not cosmetic.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Permission Is Necessary — but It Is Not Assurance\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Suppose an agent has permission to send email. Access control answers: \u003Cb>Can this agent send email?\u003C\u002Fb> It does not answer: \u003Cb>Should this particular email be sent to this particular person with this particular attachment right now?\u003C\u002Fb>\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"CAPABILITY CONTROL\\nWhat is the agent technically allowed to do? + ACTION ASSURANCE\\nIs this specific action appropriate in the current state?\"},\"type\":\"code\"},{\"data\":{\"text\":\"RBAC, OAuth scopes, API permissions and agent identities define the space of possible actions. They do not prove that an action inside that space is appropriate. Strong agent architecture needs both layers.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"The First Wrong Step Matters\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"When an agent fails, the final incorrect action is often not where the failure started. The real failure may have happened much earlier.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"Wrong retrieval ↓\\nWrong assumption ↓\\nPlausible reasoning ↓\\nValid tool call ↓\\nWrong action\"},\"type\":\"code\"},{\"data\":{\"text\":\"If we investigate only the final action, we fix the symptom. If we inspect the trajectory, we can identify the \u003Cb>first wrong step\u003C\u002Fb>. That turns an unattributable failure into a concrete engineering problem.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Agent Testing Must Move Beyond Prompt Testing\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Prompts matter, but production agent behavior emerges from an entire system.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"MODEL\\n+\\nSYSTEM PROMPT\\n+\\nCONTEXT\\n+\\nMEMORY\\n+\\nRETRIEVAL\\n+\\nTOOLS\\n+\\nPERMISSIONS\\n+\\nWORKFLOW\\n+\\nEXTERNAL STATE\\n+\\nCONTROL LOGIC\"},\"type\":\"code\"},{\"data\":{\"text\":\"Changing any one of these can change the trajectory. Therefore versioning only the prompt is insufficient.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"model_version\\nprompt_version\\ntool_version\\npolicy_version\\nretrieval_version\\nworkflow_version\\nenvironment_state\\nexecution_id\"},\"type\":\"code\"},{\"data\":{\"text\":\"Acceptance Criteria for Agents Must Include Behavior\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Traditional acceptance criteria often look like this: \u003Ci>Given X, the system produces Y.\u003C\u002Fi> For agentic systems, that is incomplete. Acceptance criteria should also define constraints on the trajectory.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Outcome\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"The customer's address is updated correctly.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Authorization\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"The agent modifies only the explicitly selected customer.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Data access\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"No unrelated customer records are accessed.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Tools\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"Only approved CRM operations are used.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Verification\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"The new address is read back and compared with the requested value.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Failure\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"Ambiguous identity resolution stops execution.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Human authority\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"A human can reject the modification before execution when risk thresholds require approval.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Evidence\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"The execution leaves a trace sufficient to reconstruct the decision and state transition.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Recovery\",\"level\":3},\"type\":\"header\"},{\"data\":{\"text\":\"The previous value remains recoverable.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Human-in-the-Loop Is Not Enough\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Adding a human approval box does not automatically solve the problem. A human can only control an agent if the person has visibility, authority, time, context and recovery capability.\"},\"type\":\"paragraph\"},{\"data\":{\"items\":[\"\u003Cb>Visibility:\u003C\u002Fb> enough information to understand what is happening.\",\"\u003Cb>Authority:\u003C\u002Fb> actual ability to stop or modify the action.\",\"\u003Cb>Time:\u003C\u002Fb> intervention before the consequence occurs.\",\"\u003Cb>Context:\u003C\u002Fb> sufficient evidence to make the decision.\",\"\u003Cb>Recovery capability:\u003C\u002Fb> ability to reverse or repair the action.\"],\"style\":\"unordered\"},\"type\":\"list\"},{\"data\":{\"text\":\"A user clicking \u003Cb>Approve\u003C\u002Fb> on something they cannot meaningfully inspect is not strong governance. It is approval theater.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Rollback Must Become a Native AI Capability\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Traditional software deployment has taught us something valuable: \u003Cb>Never deploy what you cannot roll back.\u003C\u002Fb> We should apply the same principle to agentic actions.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"REVERSIBLE\\nCan automatically undo. COMPENSATABLE\\nCannot undo directly but can execute a compensating action. IRREVERSIBLE\\nCannot reliably restore the previous state.\"},\"type\":\"code\"},{\"data\":{\"text\":\"The higher the irreversibility, the stronger the control requirement should become.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"Read public document → low consequence\\nCreate draft → reversible\\nModify CRM record → reversible but consequential\\nSend external email → practically irreversible\\nTransfer money → high consequence\\nDelete production data → potentially catastrophic\"},\"type\":\"code\"},{\"data\":{\"text\":\"The Agent Needs a Control Plane\",\"level\":2},\"type\":\"header\"},{\"data\":{\"code\":\"USER \u002F SYSTEM INTENT │ ▼ AI AGENT │ proposed action │ ▼ ┌───────────────────┐ │ CONTROL PLANE │ ├───────────────────┤ │ Identity │ │ Authorization │ │ Policy │ │ Risk │ │ State │ │ Evidence │ │ Human authority │ │ Rollback │ └───────────────────┘ │ approved? \u002F \\\\ NO YES │ │ STOP ▼ TOOL │ ▼ STATE CHANGE │ ▼ VERIFICATION\"},\"type\":\"code\"},{\"data\":{\"text\":\"\u003Cb>The LLM should propose. The control plane should govern.\u003C\u002Fb> That separation is crucial. The model should not be the ultimate authority determining whether its own proposed high-impact action is safe.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"From Benchmarks to Operational Trust\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Benchmarks remain useful. They tell us about capability, compare models, detect regressions and help estimate expected performance. But capability evaluation and operational trust answer different questions.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"A benchmark asks: \u003Cb>Can the system do this?\u003C\u002Fb> Operational assurance asks: \u003Cb>Can we allow the system to do this here, under these conditions, with these permissions and consequences?\u003C\u002Fb>\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Reliability Should Be Measured as a System Property\",\"level\":2},\"type\":\"header\"},{\"data\":{\"items\":[\"\u003Cb>Outcome correctness:\u003C\u002Fb> Did the system produce the expected result?\",\"\u003Cb>Trajectory correctness:\u003C\u002Fb> Did it follow an acceptable path?\",\"\u003Cb>Control integrity:\u003C\u002Fb> Were authorization, policy and intervention boundaries respected?\",\"\u003Cb>Recoverability:\u003C\u002Fb> Can failures be contained, reversed or repaired?\",\"\u003Cb>Evidence completeness:\u003C\u002Fb> Can the execution be reconstructed and audited?\"],\"style\":\"ordered\"},\"type\":\"list\"},{\"data\":{\"code\":\"Operational Reliability\\n=\\nOutcome × Trajectory × Control × Recoverability × Evidence\"},\"type\":\"code\"},{\"data\":{\"text\":\"The multiplication is intentional. If one critical dimension approaches zero, a high score elsewhere should not hide it. A perfectly correct output with zero authorization integrity is not an 80% reliable system. It is an unacceptable execution that happened to produce the right answer.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Success Is Sometimes the Most Dangerous Failure\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"Failures attract attention. Success often does not. That makes successful but uncontrolled agent trajectories particularly dangerous. An obvious failure creates an incident. A hidden trajectory defect creates \u003Cb>confidence\u003C\u002Fb>. And confidence expands autonomy.\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Organizations should therefore not only investigate \u003Ci>Why did the agent fail?\u003C\u002Fi> They should periodically ask: \u003Cb>Why did the agent succeed?\u003C\u002Fb> Did it succeed because the architecture reliably constrained and verified the execution, or because nothing went wrong this time?\"},\"type\":\"paragraph\"},{\"data\":{\"text\":\"Conclusion\",\"level\":2},\"type\":\"header\"},{\"data\":{\"text\":\"The industry is moving rapidly from AI that \u003Cb>answers\u003C\u002Fb> toward AI that \u003Cb>acts\u003C\u002Fb>. That transition changes what reliability means. For an answer system, evaluating the answer may often be sufficient. For an action system, we must evaluate the path.\"},\"type\":\"paragraph\"},{\"data\":{\"code\":\"Prompt ↓\\nResponse becomes Intent ↓\\nTrajectory ↓\\nActions ↓\\nState changes ↓\\nEvidence ↓\\nOutcome\"},\"type\":\"code\"},{\"data\":{\"text\":\"The final answer remains important, but it is only the visible end of a much larger system. Once AI is allowed to affect the real world, \u003Cb>the path to the answer becomes part of the answer.\u003C\u002Fb>\"},\"type\":\"paragraph\"}],\"version\":\"2.31.0\"}",{"time":212,"blocks":213,"version":560},1788955485785,[214,218,221,224,227,231,234,249,252,255,266,269,272,275,279,282,288,296,299,302,324,327,330,333,336,351,354,357,360,363,366,369,372,375,378,381,384,387,390,393,396,399,402,405,408,411,414,417,421,424,427,430,433,436,439,442,445,448,451,454,457,460,463,466,469,472,475,478,486,489,492,495,498,501,504,507,510,513,516,519,522,525,533,536,539,542,545,548,551,554,557],{"data":215,"type":217},{"text":216},"\u003Cb>Correct output does not prove correct reasoning, safe execution, or a trustworthy system.\u003C\u002Fb>","paragraph",{"data":219,"type":217},{"text":220},"For years, AI evaluation has been dominated by a deceptively simple question: \u003Cb>Was the answer correct?\u003C\u002Fb> For a chatbot, this may sometimes be sufficient. For an agent capable of searching systems, reading data, calling tools, modifying state, executing workflows, writing files, interacting with APIs, or making decisions, it is not.",{"data":222,"type":217},{"text":223},"An agent can produce the correct final answer while doing several things wrong on the way there. It can use the wrong source, misunderstand an instruction and later compensate for the mistake, access unnecessary information, execute an unauthorized intermediate action, silently recover from an error that should have triggered escalation, or leave behind side effects nobody noticed.",{"data":225,"type":217},{"text":226},"That creates one of the central problems of agentic AI: \u003Cb>a correct outcome does not prove a correct trajectory.\u003C\u002Fb>",{"data":228,"type":42},{"text":229,"level":230},"The Outcome Illusion",2,{"data":232,"type":217},{"text":233},"Traditional software gives us an intuitive model of correctness. Input enters a deterministic or mostly deterministic system, logic is executed, output is produced, and tests verify expected behavior. LLM-based systems weaken this assumption. Agentic systems go further.",{"data":235,"type":248},{"items":236,"style":247},[237,238,239,240,241,242,243,244,245,246],"model interpretation","retrieved context","tool selection","intermediate observations","external state","previous actions","model-generated plans","permission boundaries","retries and fallback behavior","human interaction","unordered","list",{"data":250,"type":217},{"text":251},"Two executions starting from nearly identical inputs may reach the same result through very different paths. If evaluation observes only the final output, most of the system remains invisible.",{"data":253,"type":217},{"text":254},"Imagine an AI agent receives the instruction: \u003Ci>Update the customer's billing address.\u003C\u002Fi> The address is ultimately updated correctly. A conventional evaluation might classify the task as successful.",{"data":256,"type":248},{"items":257,"style":265},[258,259,260,261,262,263,264],"The agent searches several unrelated customer records.","It retrieves more personal information than required.","It initially modifies the wrong account.","It notices the mistake.","It reverses the change.","It updates the correct account.","It reports success.","ordered",{"data":267,"type":217},{"text":268},"\u003Cb>Final state: correct. System behavior: unacceptable.\u003C\u002Fb> An outcome-only benchmark gives this execution a pass. A production assurance system should not.",{"data":270,"type":42},{"text":271,"level":230},"The Trajectory Is Part of the Product",{"data":273,"type":217},{"text":274},"This is why the \u003Cb>trajectory\u003C\u002Fb> of an AI agent must become a first-class engineering object. A trajectory is the sequence of relevant states and actions between the original request and the final result.",{"data":276,"type":278},{"code":277},"Intent → Context → Decision → Tool → Action → Observation → Decision → State change → Result","code",{"data":280,"type":217},{"text":281},"Zachary J. Stevens develops this idea in \u003Ci>The Trajectory Is the System\u003C\u002Fi>, arguing that agentic evaluation must move beyond the final answer and examine the complete path of action through a changing environment.",{"data":283,"type":287},{"text":284,"caption":285,"alignment":286},"A correct outcome does not excuse an unacceptable trajectory.","Zachary J. Stevens, The Trajectory Is the System","left","quote",{"data":289,"type":295},{"link":290,"meta":291},"https:\u002F\u002Fzacharyjstevens.com\u002Fdispatches\u002Fvanguard-signal\u002F009-the-trajectory-is-the-system\u002F",{"image":292,"title":293,"description":294},{},"The Trajectory Is the System","Zachary J. Stevens — DFEI.009 on evaluating agentic systems by their complete trajectory rather than only the final outcome.","linkTool",{"data":297,"type":217},{"text":298},"The distinction matters enormously. Reliability is therefore not simply \u003Cb>correct output\u003C\u002Fb>. It is closer to \u003Cb>acceptable outcome + acceptable trajectory + recoverability + evidence\u003C\u002Fb>.",{"data":300,"type":42},{"text":301,"level":230},"A Correct Answer Can Hide a Broken System",{"data":303,"type":323},{"content":304,"withHeadings":14},[305,309,313,316,320],[306,307,308],"Agent","Final result","Execution",[310,311,312],"A","Correct","Correct path",[314,311,315],"B","Unsafe path",[317,318,319],"C","Incorrect","Safe failure",[321,318,322],"D","Unsafe failure","table",{"data":325,"type":217},{"text":326},"Most benchmark-driven evaluation strongly rewards A and B and penalizes C and D. Operationally, however, \u003Cb>B can be more dangerous than C\u003C\u002Fb>. Agent C may recognize uncertainty, stop execution and request human review. Agent B may confidently produce correct results while violating assumptions that nobody is monitoring.",{"data":328,"type":278},{"code":329},"successful output → increased trust → broader permissions → more automation → larger blast radius",{"data":331,"type":42},{"text":332,"level":230},"We Need Evidence, Not Confidence",{"data":334,"type":217},{"text":335},"One of the biggest mistakes in AI adoption is treating model confidence, user satisfaction or historical success rate as evidence of system reliability. They are not equivalent.",{"data":337,"type":248},{"items":338,"style":247},[339,340,341,342,343,344,345,346,347,348,349,350],"What did the agent receive?","What context did it retrieve?","Which tools did it call?","Why was the action allowed?","What state existed before the action?","What changed?","Which intermediate failures occurred?","Was anything retried?","Was human approval required?","Could execution have been stopped?","Can the action be reversed?","Which model, prompt and tool versions were involved?",{"data":352,"type":217},{"text":353},"Without these answers, there is no serious operational assurance. There is only an output. Observability and evidence must therefore be designed into agent architecture rather than added after deployment.",{"data":355,"type":42},{"text":356,"level":230},"Logging Is Not the Same as Control",{"data":358,"type":217},{"text":359},"Organizations often respond: \u003Ci>Everything is logged.\u003C\u002Fi> Good. But logging alone does not control anything. A log tells you what happened. A control determines whether something \u003Cb>may happen\u003C\u002Fb>.",{"data":361,"type":278},{"code":362},"Agent requests DELETE \u002Fcustomer\u002F123 ↓\nAction logged ↓\nDELETE executed",{"data":364,"type":217},{"text":365},"That gives observability. Compare it with:",{"data":367,"type":278},{"code":368},"Agent requests DELETE \u002Fcustomer\u002F123 ↓\nPolicy evaluation ↓\nCurrent identity verified ↓\nCurrent action parameters checked ↓\nRisk threshold evaluated ↓\nHuman approval if required ↓\nAction executed ↓\nResult verified ↓\nEvidence stored",{"data":370,"type":217},{"text":371},"Now we are approaching a control system. The difference is architectural, not cosmetic.",{"data":373,"type":42},{"text":374,"level":230},"Permission Is Necessary — but It Is Not Assurance",{"data":376,"type":217},{"text":377},"Suppose an agent has permission to send email. Access control answers: \u003Cb>Can this agent send email?\u003C\u002Fb> It does not answer: \u003Cb>Should this particular email be sent to this particular person with this particular attachment right now?\u003C\u002Fb>",{"data":379,"type":278},{"code":380},"CAPABILITY CONTROL\nWhat is the agent technically allowed to do? + ACTION ASSURANCE\nIs this specific action appropriate in the current state?",{"data":382,"type":217},{"text":383},"RBAC, OAuth scopes, API permissions and agent identities define the space of possible actions. They do not prove that an action inside that space is appropriate. Strong agent architecture needs both layers.",{"data":385,"type":42},{"text":386,"level":230},"The First Wrong Step Matters",{"data":388,"type":217},{"text":389},"When an agent fails, the final incorrect action is often not where the failure started. The real failure may have happened much earlier.",{"data":391,"type":278},{"code":392},"Wrong retrieval ↓\nWrong assumption ↓\nPlausible reasoning ↓\nValid tool call ↓\nWrong action",{"data":394,"type":217},{"text":395},"If we investigate only the final action, we fix the symptom. If we inspect the trajectory, we can identify the \u003Cb>first wrong step\u003C\u002Fb>. That turns an unattributable failure into a concrete engineering problem.",{"data":397,"type":42},{"text":398,"level":230},"Agent Testing Must Move Beyond Prompt Testing",{"data":400,"type":217},{"text":401},"Prompts matter, but production agent behavior emerges from an entire system.",{"data":403,"type":278},{"code":404},"MODEL\n+\nSYSTEM PROMPT\n+\nCONTEXT\n+\nMEMORY\n+\nRETRIEVAL\n+\nTOOLS\n+\nPERMISSIONS\n+\nWORKFLOW\n+\nEXTERNAL STATE\n+\nCONTROL LOGIC",{"data":406,"type":217},{"text":407},"Changing any one of these can change the trajectory. Therefore versioning only the prompt is insufficient.",{"data":409,"type":278},{"code":410},"model_version\nprompt_version\ntool_version\npolicy_version\nretrieval_version\nworkflow_version\nenvironment_state\nexecution_id",{"data":412,"type":42},{"text":413,"level":230},"Acceptance Criteria for Agents Must Include Behavior",{"data":415,"type":217},{"text":416},"Traditional acceptance criteria often look like this: \u003Ci>Given X, the system produces Y.\u003C\u002Fi> For agentic systems, that is incomplete. Acceptance criteria should also define constraints on the trajectory.",{"data":418,"type":42},{"text":419,"level":420},"Outcome",3,{"data":422,"type":217},{"text":423},"The customer's address is updated correctly.",{"data":425,"type":42},{"text":426,"level":420},"Authorization",{"data":428,"type":217},{"text":429},"The agent modifies only the explicitly selected customer.",{"data":431,"type":42},{"text":432,"level":420},"Data access",{"data":434,"type":217},{"text":435},"No unrelated customer records are accessed.",{"data":437,"type":42},{"text":438,"level":420},"Tools",{"data":440,"type":217},{"text":441},"Only approved CRM operations are used.",{"data":443,"type":42},{"text":444,"level":420},"Verification",{"data":446,"type":217},{"text":447},"The new address is read back and compared with the requested value.",{"data":449,"type":42},{"text":450,"level":420},"Failure",{"data":452,"type":217},{"text":453},"Ambiguous identity resolution stops execution.",{"data":455,"type":42},{"text":456,"level":420},"Human authority",{"data":458,"type":217},{"text":459},"A human can reject the modification before execution when risk thresholds require approval.",{"data":461,"type":42},{"text":462,"level":420},"Evidence",{"data":464,"type":217},{"text":465},"The execution leaves a trace sufficient to reconstruct the decision and state transition.",{"data":467,"type":42},{"text":468,"level":420},"Recovery",{"data":470,"type":217},{"text":471},"The previous value remains recoverable.",{"data":473,"type":42},{"text":474,"level":230},"Human-in-the-Loop Is Not Enough",{"data":476,"type":217},{"text":477},"Adding a human approval box does not automatically solve the problem. A human can only control an agent if the person has visibility, authority, time, context and recovery capability.",{"data":479,"type":248},{"items":480,"style":247},[481,482,483,484,485],"\u003Cb>Visibility:\u003C\u002Fb> enough information to understand what is happening.","\u003Cb>Authority:\u003C\u002Fb> actual ability to stop or modify the action.","\u003Cb>Time:\u003C\u002Fb> intervention before the consequence occurs.","\u003Cb>Context:\u003C\u002Fb> sufficient evidence to make the decision.","\u003Cb>Recovery capability:\u003C\u002Fb> ability to reverse or repair the action.",{"data":487,"type":217},{"text":488},"A user clicking \u003Cb>Approve\u003C\u002Fb> on something they cannot meaningfully inspect is not strong governance. It is approval theater.",{"data":490,"type":42},{"text":491,"level":230},"Rollback Must Become a Native AI Capability",{"data":493,"type":217},{"text":494},"Traditional software deployment has taught us something valuable: \u003Cb>Never deploy what you cannot roll back.\u003C\u002Fb> We should apply the same principle to agentic actions.",{"data":496,"type":278},{"code":497},"REVERSIBLE\nCan automatically undo. COMPENSATABLE\nCannot undo directly but can execute a compensating action. IRREVERSIBLE\nCannot reliably restore the previous state.",{"data":499,"type":217},{"text":500},"The higher the irreversibility, the stronger the control requirement should become.",{"data":502,"type":278},{"code":503},"Read public document → low consequence\nCreate draft → reversible\nModify CRM record → reversible but consequential\nSend external email → practically irreversible\nTransfer money → high consequence\nDelete production data → potentially catastrophic",{"data":505,"type":42},{"text":506,"level":230},"The Agent Needs a Control Plane",{"data":508,"type":278},{"code":509},"USER \u002F SYSTEM INTENT │ ▼ AI AGENT │ proposed action │ ▼ ┌───────────────────┐ │ CONTROL PLANE │ ├───────────────────┤ │ Identity │ │ Authorization │ │ Policy │ │ Risk │ │ State │ │ Evidence │ │ Human authority │ │ Rollback │ └───────────────────┘ │ approved? \u002F \\ NO YES │ │ STOP ▼ TOOL │ ▼ STATE CHANGE │ ▼ VERIFICATION",{"data":511,"type":217},{"text":512},"\u003Cb>The LLM should propose. The control plane should govern.\u003C\u002Fb> That separation is crucial. The model should not be the ultimate authority determining whether its own proposed high-impact action is safe.",{"data":514,"type":42},{"text":515,"level":230},"From Benchmarks to Operational Trust",{"data":517,"type":217},{"text":518},"Benchmarks remain useful. They tell us about capability, compare models, detect regressions and help estimate expected performance. But capability evaluation and operational trust answer different questions.",{"data":520,"type":217},{"text":521},"A benchmark asks: \u003Cb>Can the system do this?\u003C\u002Fb> Operational assurance asks: \u003Cb>Can we allow the system to do this here, under these conditions, with these permissions and consequences?\u003C\u002Fb>",{"data":523,"type":42},{"text":524,"level":230},"Reliability Should Be Measured as a System Property",{"data":526,"type":248},{"items":527,"style":265},[528,529,530,531,532],"\u003Cb>Outcome correctness:\u003C\u002Fb> Did the system produce the expected result?","\u003Cb>Trajectory correctness:\u003C\u002Fb> Did it follow an acceptable path?","\u003Cb>Control integrity:\u003C\u002Fb> Were authorization, policy and intervention boundaries respected?","\u003Cb>Recoverability:\u003C\u002Fb> Can failures be contained, reversed or repaired?","\u003Cb>Evidence completeness:\u003C\u002Fb> Can the execution be reconstructed and audited?",{"data":534,"type":278},{"code":535},"Operational Reliability\n=\nOutcome × Trajectory × Control × Recoverability × Evidence",{"data":537,"type":217},{"text":538},"The multiplication is intentional. If one critical dimension approaches zero, a high score elsewhere should not hide it. A perfectly correct output with zero authorization integrity is not an 80% reliable system. It is an unacceptable execution that happened to produce the right answer.",{"data":540,"type":42},{"text":541,"level":230},"Success Is Sometimes the Most Dangerous Failure",{"data":543,"type":217},{"text":544},"Failures attract attention. Success often does not. That makes successful but uncontrolled agent trajectories particularly dangerous. An obvious failure creates an incident. A hidden trajectory defect creates \u003Cb>confidence\u003C\u002Fb>. And confidence expands autonomy.",{"data":546,"type":217},{"text":547},"Organizations should therefore not only investigate \u003Ci>Why did the agent fail?\u003C\u002Fi> They should periodically ask: \u003Cb>Why did the agent succeed?\u003C\u002Fb> Did it succeed because the architecture reliably constrained and verified the execution, or because nothing went wrong this time?",{"data":549,"type":42},{"text":550,"level":230},"Conclusion",{"data":552,"type":217},{"text":553},"The industry is moving rapidly from AI that \u003Cb>answers\u003C\u002Fb> toward AI that \u003Cb>acts\u003C\u002Fb>. That transition changes what reliability means. For an answer system, evaluating the answer may often be sufficient. For an action system, we must evaluate the path.",{"data":555,"type":278},{"code":556},"Prompt ↓\nResponse becomes Intent ↓\nTrajectory ↓\nActions ↓\nState changes ↓\nEvidence ↓\nOutcome",{"data":558,"type":217},{"text":559},"The final answer remains important, but it is only the visible end of a much larger system. Once AI is allowed to affect the real world, \u003Cb>the path to the answer becomes part of the answer.\u003C\u002Fb>","2.31.0","Correct output does not prove correct reasoning, safe execution, or a trustworthy system.","\u002Fuploads\u002F2026\u002F09\u002Fai-agent-reliability-why-the-final-answer-is-not-enough-1788955466306-pl0qhz.webp","ai-agent-reliability-why-the-final-answer-is-not-enough-1788955466306-pl0qhz","PUBLISHED","2026-09-09T04:01:00.000Z","2026-09-09T12:01:07.219Z","2026-09-09T13:08:53.481Z",{"en":569,"de":570,"sr":571,"es":572,"fr":573,"it":574,"ru":575,"zh":576},"\u002Fblog\u002Fai-agent-reliability-why-the-final-answer-is-not-enough","\u002Fde\u002Fblog\u002Fai-agent-reliability-why-the-final-answer-is-not-enough","\u002Fsr\u002Fblog\u002Fai-agent-reliability-why-the-final-answer-is-not-enough","\u002Fes\u002Fblog\u002Fai-agent-reliability-why-the-final-answer-is-not-enough","\u002Ffr\u002Fblog\u002Fai-agent-reliability-why-the-final-answer-is-not-enough","\u002Fit\u002Fblog\u002Fai-agent-reliability-why-the-final-answer-is-not-enough","\u002Fru\u002Fblog\u002Fai-agent-reliability-why-the-final-answer-is-not-enough","\u002Fzh\u002Fblog\u002Fai-agent-reliability-why-the-final-answer-is-not-enough",[578,582],{"id":579,"name":580,"slug":581},58,"Evaluation & Quality Gates","evaluation",{"id":583,"name":584,"slug":585},59,"Governance & Auditability","governance",{"id":587,"login":588,"email":589,"displayName":590},"20","rooth8233","aleksandar@stajic.de","Aleksandar Stajić",[592],{"lang":7,"title":208,"content":210,"contentJson":593,"excerpt":561},{"time":212,"blocks":594,"version":560},[595,597,599,601,603,605,607,610,612,614,617,619,621,623,625,627,629,633,635,637,645,647,649,651,653,656,658,660,662,664,666,668,670,672,674,676,678,680,682,684,686,688,690,692,694,696,698,700,702,704,706,708,710,712,714,716,718,720,722,724,726,728,730,732,734,736,738,740,743,745,747,749,751,753,755,757,759,761,763,765,767,769,772,774,776,778,780,782,784,786,788],{"data":596,"type":217},{"text":216},{"data":598,"type":217},{"text":220},{"data":600,"type":217},{"text":223},{"data":602,"type":217},{"text":226},{"data":604,"type":42},{"text":229,"level":230},{"data":606,"type":217},{"text":233},{"data":608,"type":248},{"items":609,"style":247},[237,238,239,240,241,242,243,244,245,246],{"data":611,"type":217},{"text":251},{"data":613,"type":217},{"text":254},{"data":615,"type":248},{"items":616,"style":265},[258,259,260,261,262,263,264],{"data":618,"type":217},{"text":268},{"data":620,"type":42},{"text":271,"level":230},{"data":622,"type":217},{"text":274},{"data":624,"type":278},{"code":277},{"data":626,"type":217},{"text":281},{"data":628,"type":287},{"text":284,"caption":285,"alignment":286},{"data":630,"type":295},{"link":290,"meta":631},{"image":632,"title":293,"description":294},{},{"data":634,"type":217},{"text":298},{"data":636,"type":42},{"text":301,"level":230},{"data":638,"type":323},{"content":639,"withHeadings":14},[640,641,642,643,644],[306,307,308],[310,311,312],[314,311,315],[317,318,319],[321,318,322],{"data":646,"type":217},{"text":326},{"data":648,"type":278},{"code":329},{"data":650,"type":42},{"text":332,"level":230},{"data":652,"type":217},{"text":335},{"data":654,"type":248},{"items":655,"style":247},[339,340,341,342,343,344,345,346,347,348,349,350],{"data":657,"type":217},{"text":353},{"data":659,"type":42},{"text":356,"level":230},{"data":661,"type":217},{"text":359},{"data":663,"type":278},{"code":362},{"data":665,"type":217},{"text":365},{"data":667,"type":278},{"code":368},{"data":669,"type":217},{"text":371},{"data":671,"type":42},{"text":374,"level":230},{"data":673,"type":217},{"text":377},{"data":675,"type":278},{"code":380},{"data":677,"type":217},{"text":383},{"data":679,"type":42},{"text":386,"level":230},{"data":681,"type":217},{"text":389},{"data":683,"type":278},{"code":392},{"data":685,"type":217},{"text":395},{"data":687,"type":42},{"text":398,"level":230},{"data":689,"type":217},{"text":401},{"data":691,"type":278},{"code":404},{"data":693,"type":217},{"text":407},{"data":695,"type":278},{"code":410},{"data":697,"type":42},{"text":413,"level":230},{"data":699,"type":217},{"text":416},{"data":701,"type":42},{"text":419,"level":420},{"data":703,"type":217},{"text":423},{"data":705,"type":42},{"text":426,"level":420},{"data":707,"type":217},{"text":429},{"data":709,"type":42},{"text":432,"level":420},{"data":711,"type":217},{"text":435},{"data":713,"type":42},{"text":438,"level":420},{"data":715,"type":217},{"text":441},{"data":717,"type":42},{"text":444,"level":420},{"data":719,"type":217},{"text":447},{"data":721,"type":42},{"text":450,"level":420},{"data":723,"type":217},{"text":453},{"data":725,"type":42},{"text":456,"level":420},{"data":727,"type":217},{"text":459},{"data":729,"type":42},{"text":462,"level":420},{"data":731,"type":217},{"text":465},{"data":733,"type":42},{"text":468,"level":420},{"data":735,"type":217},{"text":471},{"data":737,"type":42},{"text":474,"level":230},{"data":739,"type":217},{"text":477},{"data":741,"type":248},{"items":742,"style":247},[481,482,483,484,485],{"data":744,"type":217},{"text":488},{"data":746,"type":42},{"text":491,"level":230},{"data":748,"type":217},{"text":494},{"data":750,"type":278},{"code":497},{"data":752,"type":217},{"text":500},{"data":754,"type":278},{"code":503},{"data":756,"type":42},{"text":506,"level":230},{"data":758,"type":278},{"code":509},{"data":760,"type":217},{"text":512},{"data":762,"type":42},{"text":515,"level":230},{"data":764,"type":217},{"text":518},{"data":766,"type":217},{"text":521},{"data":768,"type":42},{"text":524,"level":230},{"data":770,"type":248},{"items":771,"style":265},[528,529,530,531,532],{"data":773,"type":278},{"code":535},{"data":775,"type":217},{"text":538},{"data":777,"type":42},{"text":541,"level":230},{"data":779,"type":217},{"text":544},{"data":781,"type":217},{"text":547},{"data":783,"type":42},{"text":550,"level":230},{"data":785,"type":217},{"text":553},{"data":787,"type":278},{"code":556},{"data":789,"type":217},{"text":559},"Post erfolgreich abgerufen",{"items":792,"source":807,"manualIds":808,"manualMatchedIds":809},[793,800],{"id":794,"slug":795,"title":796,"excerpt":797,"featuredImage":798,"publishedAt":799},"459","ollama-is-not-the-product-building-production-ready-open-llm-applications","Ollama Is Not the Product: Building Production-Ready Open-LLM Applications","Running a local model with Ollama is easy. Building a production-ready Open-LLM application is harder: it requires RAG, access control, provider abstraction, evaluation, logging, deployment discipline and a controlled application layer around the model.\n","\u002Fuploads\u002F2026\u002F06\u002Follama-is-not-the-product-building-production-ready-open-llm-applications-1782679361640-h0usqf.webp","2026-06-28T16:39:00.000Z",{"id":801,"slug":802,"title":803,"excerpt":804,"featuredImage":805,"publishedAt":806},"364","tipps-fuer-die-verbesserung-der-seo-suchmaschinenoptimierung","Mastering the SEO Workflow: Essential Optimization Strategies for Organic Growth","A structured SEO workflow is crucial for sustainable organic growth. Learn the ten foundational strategies, from keyword research and technical optimization to content quality and performance analysis.","\u002Fuploads\u002F2026\u002F03\u002Ftipps-fuer-die-verbesserung-der-seo-suchmaschinenoptimierung-1774866098131-hwkzrg.webp","2024-01-26T06:35:00.000Z","fallback",[],[]]