{"templateName":"blog-page","cssClassNames":"blog-page page basicpage summit-page","allowedRenditionsWidth":["320","480","640","768","960","1200","1440","1920"],"description":"Learn how new Cortex Agents capabilities simplify building, deploying, and operating enterprise AI agents with managed orchestration, execution, and governance.","language":"en","title":"What's New in Snowflake Cortex Agents for Enterprise AI","analyticsPageType":"homepage","analyticsCategory":"general","analyticsSubCategory":"","excludeFromAnalytics":false,":items":{"root":{"columnClassNames":{"experiencefragment-banner":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-sub-header":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-pre-footer":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-header":"aem-GridColumn aem-GridColumn--default--12","markup_editor-table":"aem-GridColumn aem-GridColumn--default--12","responsivegrid":"aem-GridColumn aem-GridColumn--default--12","experiencefragment-footer":"aem-GridColumn aem-GridColumn--default--12","markup_editor":"aem-GridColumn aem-GridColumn--default--12","container_47873732":"aem-GridColumn aem-GridColumn--default--12"},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12",":items":{"experiencefragment-banner":{"id":"experiencefragment-dd49b465bd","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/pushdown-banner/pushdown-banner-blank/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/pushdown-banner/pushdown-banner-blank.xfmodel.json"},"experiencefragment-header":{"id":"experiencefragment-f27363df49","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/mega-nav-header/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/mega-nav-header/master.xfmodel.json","languageNavPath":"/content/snowflake-site/global/en/blog/snowflake-cortex-agents-enterprise-ai-scale.languagenav.json","appliedCssClassNames":"snowflake-sticky-nav-host"},"experiencefragment-sub-header":{"id":"experiencefragment-876da4d15b","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/sub-navigation/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/sub-navigation/master.xfmodel.json"},"responsivegrid":{"columnClassNames":{"container_breadcrumb":"aem-GridColumn aem-GridColumn--default--12","container_main_content":"aem-GridColumn aem-GridColumn--default--12"},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12",":items":{"container_breadcrumb":{"layout":"RESPONSIVE_GRID","columnClassNames":{"breadcrumb":"aem-GridColumn aem-GridColumn--default--12"},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","id":"blog-page-breadcrumb-indentation",":items":{"breadcrumb":{"id":"breadcrumb-03bb663f7d","breadcrumbItems":[{"title":"Blog","path":"/en/blog/","active":false},{"title":"AI & ML","path":"/en/blog/ai-ml/","active":false},{"title":"Your Agents Are in Production. Now What?","path":"/en/blog/snowflake-cortex-agents-enterprise-ai-scale/","active":false}],":type":"snowflake-site/components/blog/breadcrumb"}},":itemsOrder":["breadcrumb"],"appliedCssClassNames":"snowflake-container",":type":"snowflake-site/components/container"},"container_main_content":{"layout":"RESPONSIVE_GRID","columnClassNames":{"flexible_column_container":"aem-GridColumn aem-GridColumn--default--12","related_content":"aem-GridColumn aem-GridColumn--default--12"},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","id":"main-content",":items":{"flexible_column_container":{"id":"flexible-column-container-1bebc53fc4","propertiesId":"snowflake-blog-template-main-container","type":"2-column-60-40","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"none","bottomPadding":"none","spaceBetween":"none","reverseOnMobile":true,"carouselOnMobile":false,"backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-899799fda8",":items":{"container_hero":{"layout":"RESPONSIVE_GRID","columnClassNames":{"blog_hero":"aem-GridColumn aem-GridColumn--default--12"},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","id":"container-c219224c49",":items":{"blog_hero":{"id":"blog-hero-f6857dd620","linkedInShareUrl":"https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwww.snowflake.com%2Fcontent%2Fsnowflake-site%2Fglobal%2Fen%2Fblog%2Fsnowflake-cortex-agents-enterprise-ai-scale&title=Your+Agents+Are+in+Production.+Now+What%3F","twitterShareUrl":"https://x.com/intent/post?url=https%3A%2F%2Fwww.snowflake.com%2Fcontent%2Fsnowflake-site%2Fglobal%2Fen%2Fblog%2Fsnowflake-cortex-agents-enterprise-ai-scale&text=Your+Agents+Are+in+Production.+Now+What%3F","facebookShareUrl":"https://www.facebook.com/sharer/sharer.php?u=https%3A%2F%2Fwww.snowflake.com%2Fcontent%2Fsnowflake-site%2Fglobal%2Fen%2Fblog%2Fsnowflake-cortex-agents-enterprise-ai-scale","showClaude":true,"showChatGpt":true,"authors":[{"authorImage":{"id":"image-152cab9b6c","height":"653","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--8049a19c-6f69-400d-9f50-c2ee72ba1d88/arun-agarwal.jpg?preferwebp=true&quality=85","alt":"Arun Agarwal","width":"720",":type":"snowflake-site/components/image"},"authorCta":{"id":"button-e3d1624beb","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/arun-agarwal/"},"linkTargetContentType":"DOCUMENT_LEARN",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Arun Agarwal"}},{"authorImage":{"id":"image-8e7bdd9fae","height":"400","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--bab3c432-9d00-4fbb-9785-e632873b6c56/harshal.jpg?preferwebp=true&quality=85","alt":"Harshal Pimpalkhute","width":"400",":type":"snowflake-site/components/image"},"authorCta":{"id":"button-0cf52a41a9","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/harshal-pimpalkhute/"},"linkTargetContentType":"DOCUMENT_LEARN",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Harshal Pimpalkhute"}},{"authorImage":{"id":"image-b93ca589f5","height":"2549","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--3ab8a4fd-dd09-4fd7-b945-2d43de0f11d8/william-allen.jpg?preferwebp=true&quality=85","width":"2549",":type":"snowflake-site/components/image"},"authorCta":{"id":"button-fa50caa4f2","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/william-allen/"},"linkTargetContentType":"DOCUMENT_LEARN",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"William Allen"}},{"authorImage":{"id":"image-7a5c55a5fb","height":"346","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--67ef09e9-67c2-4577-b469-c3cc5c2b457a/vino-duraisamy.png?preferwebp=true&quality=85","width":"346",":type":"snowflake-site/components/image"},"authorCta":{"id":"button-31b2407b6c","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/vino-duraisamy/"},"linkTargetContentType":"DOCUMENT_LEARN",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Vino Duraisamy"}}],"image":{"id":"image-deafd7cfa6","height":"720","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--4eb7d888-b741-43ef-a174-bf6eaad5442b/30-snowflake.png?preferwebp=true&quality=85","width":"1680",":type":"snowflake-site/components/image"},"timeToRead":"10","publicationDate":"JUL 21, 2026","tag":{"tagText":"AI & ML","tagColor":"#C54985"},"title":{"lines":["Your Agents Are in Production. Now What?"],"type":"heading2",":type":"snowflake-site/components/title-v2"},":type":"snowflake-site/components/blog/blog-hero"}},":itemsOrder":["blog_hero"],":type":"snowflake-site/components/container"},"responsivegrid_content":{"columnClassNames":{"image":"aem-GridColumn aem-GridColumn--default--12","blog_text_1859485776":"aem-GridColumn aem-GridColumn--default--12","blog_text":"aem-GridColumn aem-GridColumn--default--12","blog_text_1528741199":"aem-GridColumn aem-GridColumn--default--12","image_1660099645":"aem-GridColumn aem-GridColumn--default--12"},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","appliedCssClassNames":"snowflake-layout-container-inner-padding-small",":items":{"blog_text":{"id":"blog-text-5a7786bf2a","text":"\u003Cp\u003EThe first generation of enterprise AI focused on building agents. The next generation is about operating them at scale.\u003C/p\u003E\r\n\u003Cp\u003E\u003Ca rel=\"noopener noreferrer\" target=\"_blank\" href=\"https://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents\"\u003ECortex Agents\u003C/a\u003E have always provided a managed runtime for building AI applications inside Snowflake. Over the past year, we've watched customers move from deploying their first agent to managing enterprise deployments. That shift has surfaced a new set of operational challenges around orchestration, execution and governance.\u003C/p\u003E\r\n\u003Cp\u003EToday, we're expanding Cortex Agents with new capabilities that simplify the operational complexity of building, deploying, and operating enterprise AI agents at scale.\u003C/p\u003E\r\n\u003Ch2\u003EAt a glance\u003C/h2\u003E\r\n\u003Cul\u003E\r\n\u003Cli\u003E\u003Cb\u003EBuild\u003C/b\u003E\u003Cul\u003E\r\n\u003Cli\u003E\u003Cb\u003ECoding Agent (in public preview soon):\u003C/b\u003E Build a managed coding agent backed by the same runtime that powers CoCo and deploy it in any application.\u003C/li\u003E\r\n\u003Cli\u003E\u003Cb\u003ESkills Package (in public preview soon):\u003C/b\u003E Package a curated set of skills into a single object and reference them in your agent with a single URI\u003C/li\u003E\r\n\u003Cli\u003E\u003Cb\u003EAgent Toolset (in public preview soon):\u003C/b\u003E Reference another agent's tools in your current agent specification instead of rewriting and maintaining duplicate tool definitions.\u003C/li\u003E\r\n\u003Cli\u003E\u003Cb\u003ETool Search (in public preview soon):\u003C/b\u003E Discover tools on demand instead of loading every definition upfront.\u003C/li\u003E\r\n\u003C/ul\u003E\r\n\u003C/li\u003E\r\n\u003C/ul\u003E\r\n\u003Cul\u003E\r\n\u003Cli\u003E\u003Cb\u003ERun\u003C/b\u003E\u003Cul\u003E\r\n\u003Cli\u003E\u003Cb\u003EAsync Agent API (generally available soon):\u003C/b\u003E Run long jobs in the background without holding a connection open.\u003C/li\u003E\r\n\u003Cli\u003E\u003Cb\u003ECode Execution Tool (in public preview soon):\u003C/b\u003E Process data and generate polished outputs inside a sandboxed Python environment.\u003C/li\u003E\r\n\u003Cli\u003E\u003Cb\u003EInterrupt and Resume (generally available soon):\u003C/b\u003E Stop a running agent, correct course and pick up where it left off.\u003C/li\u003E\r\n\u003Cli\u003E\u003Cb\u003EPartial Access (in public preview soon):\u003C/b\u003E Serve users with different permission levels from a single managed agent.\u003C/li\u003E\r\n\u003C/ul\u003E\r\n\u003C/li\u003E\r\n\u003C/ul\u003E\r\n\u003Cul\u003E\r\n\u003Cli\u003E\u003Cb\u003EManage\u003C/b\u003E\u003Cul\u003E\r\n\u003Cli\u003E\u003Cb\u003EVersioning UI (in public preview):\u003C/b\u003E Compare, roll back and promote agent configurations without editing YAML.\u003C/li\u003E\r\n\u003C/ul\u003E\r\n\u003C/li\u003E\r\n\u003C/ul\u003E\r\n\u003Ch2\u003ECortex Agents — a refresher\u003C/h2\u003E\r\n\u003Cp\u003EA Cortex Agent can reason over your enterprise data, invoke tools, execute code and take actions across connected systems — all within Snowflake's governance boundary. Every action is subject to the same access controls and audit trail that cover the rest of your data.\u003C/p\u003E\r\n\u003Cp\u003EYou define a Cortex Agent declaratively that includes its orchestration model, tools and instructions. Applications invoke it through a REST API, while Snowflake handles model routing, tool invocation and result assembly.\u003C/p\u003E\r\n\u003Ch2\u003EThe production challenge\u003C/h2\u003E\r\n\u003Cp\u003EMost deployments start with a single agent, a handful of tools and short tasks. That setup works fine for a proof of concept and maybe a small deployment.\u003C/p\u003E\r\n\u003Cp\u003EEnterprise production at scale looks different: Workflows span multiple specialized agents, tool libraries grow to dozens of integrations, requests run for minutes rather than milliseconds and governance teams need visibility into exactly what each agent is authorized to do. These aren't problems that better prompting or a faster model will solve. They're operational problems, and they're what our new capabilities below are designed to address.\u003C/p\u003E\r\n\u003Ch2\u003EBuild your agent\u003C/h2\u003E\r\n\u003Ch4\u003EBring Snowflake CoCo's coding capabilities to any application\u003C/h4\u003E\r\n\u003Cp\u003ESnowflake CoCo is the first truly Data-native AI coding agent, built to speed time to value across modern data stack. Deeply aware of your enterprise's data, compute, governance, and operational semantics, CoCo turns complex data engineering, analytics, machine learning, and agent-building tasks into simple and informed conversations with high accuracy and trust.\u003C/p\u003E\r\n\u003Cp\u003EWith the Cortex Agents REST API, you can now build a managed agent that brings those same \u003Cb\u003Ecoding capabilities (in public preview soon)\u003C/b\u003E into your own applications and interfaces. Configure the agent with the CoCo toolset (including sandbox), scope its instructions and data grounding to your environment, and expose it over a standard REST endpoint. Any query that requires code generation, SQL execution, data transformation or pipeline automation is handled inside the sandboxed runtime — the same foundation that powers CoCo, now composable as part of your own agentic workflows.\u003C/p\u003E\r\n\u003Cp\u003EFor example, our engineering team built "Debug with CoCo," an internal tool that root-causes any Cortex Agent request, in under 20 min. The app itself is a few hundred lines long: it takes a request_id via UI, and makes a single Cortex Agent API call with a dedicated prompt. The managed API provisions a CoCo sandbox, mounts the logs as the agent's grounding, runs bash and SQL through the CoCo toolset, and streams an evidenced root-cause analysis back to the browser. No coding-agent runtime, and no data leaving Snowflake — just REST calls:\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/blog/blog-text"},"image":{"id":"image-0be68239db","height":"1068","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--e37e7cfb-c278-47ce-959e-a5d7e92f6dc0/agent-request-debugger.png?preferwebp=true&quality=85","alt":"agent request debugger","width":"1820",":type":"snowflake-site/components/image"},"blog_text_1859485776":{"id":"blog-text-b29221ff30","text":"\u003Cp\u003E\u003Csub\u003E\u003Ci\u003EFigure 1:  "Debug with CoCo," an engineer tool built on the Managed Agent API with CoCo runtime.\u003C/i\u003E\u003C/sub\u003E\u003C/p\u003E\r\n\u003Ch4\u003EGive your agents a shared capability layer\u003C/h4\u003E\r\n\u003Cp\u003EAs agent deployments grow, skill management becomes a bottleneck. Every new agent needs its own list of capabilities, every update needs to be applied across each one individually, and as skill libraries expand, agent specifications become harder to maintain and easier to get out of sync.\u003C/p\u003E\r\n\u003Cp\u003E\u003Cb\u003ESkills Package (in public preview soon)\u003C/b\u003E lets you define a curated set of skills once and reference the entire collection in any agent specification with a single URI. When the package is updated, every agent that references it picks up the change automatically. Teams can build a shared capability layer — a standard set of data workflows, domain-specific tools, or approved integrations — and compose new agents on top of it without duplicating or maintaining separate copies.\u003C/p\u003E\r\n\u003Ch4\u003EReuse tools across agents without rewriting them\u003C/h4\u003E\r\n\u003Cp\u003EEnterprise agent deployments often start with a well-defined agent — one with a carefully curated set of tools, tested integrations and tuned configurations. When a second agent needs some of those same tools, the current approach is to duplicate the tool definitions: copy the spec, maintain two copies and keep them in sync whenever something changes.\u003C/p\u003E\r\n\u003Cp\u003E\u003Cb\u003EAgent Toolset (in public preview soon)\u003C/b\u003E lets you reference another agent's tools directly in your current agent's specification. A single line points to the source agent, and all its tools are available immediately — no redefinition required. When the source agent's tools are updated, every agent referencing that toolset inherits the change automatically. Teams can define a canonical set of tools once — a shared data access layer, a standard set of search integrations, a common MCP configuration — and compose new agents on top of it without duplicating or diverging from the original.\u003C/p\u003E\r\n\u003Ch4\u003EOptimize context building for your agents\u003C/h4\u003E\r\n\u003Cp\u003EA large enterprise deployment with connections to GitHub, Slack, Jira, Salesforce and internal MCP servers can easily accumulate dozens of tools. Before answering a single question, an agent may spend tens of thousands of tokens loading tool definitions into context.\u003C/p\u003E\r\n\u003Cp\u003E\u003Cb\u003ETool Search (in public preview soon)\u003C/b\u003E will address this through progressive disclosure. Instead of loading every tool definition upfront, the agent will search its tool library at inference time and will load only the definitions relevant to the current task. At scale, this will be more than a performance optimization — it can improve accuracy by reducing context window pressure and will enable agents to operate across tool ecosystems that would otherwise be impractical to load all at once.\u003C/p\u003E\r\n\u003Ch2\u003ERun your agent\u003C/h2\u003E\r\n\u003Ch4\u003ERunning long tasks shouldn't block your app\u003C/h4\u003E\r\n\u003Cp\u003ESome tasks take minutes, not milliseconds. Reconciling monthly financial data, generating a comprehensive research report, processing a batch of customer contracts — these aren't jobs where you can hold an HTTP connection open.\u003C/p\u003E\r\n\u003Cp\u003EThe \u003Cb\u003EAsync Agent API (generally available soon)\u003C/b\u003E lets agents run these tasks in the background. Callers receive a run_id immediately and retrieve responses when the task is complete.\u003C/p\u003E\r\n\u003Cp\u003EThe agent runs to completion — or pauses if it requires user input — and the result is available when ready. Long-running agent workflows can now execute entirely within the managed runtime, without requiring custom queuing or orchestration infrastructure.\u003C/p\u003E\r\n\u003Ch4\u003EFrom raw data to polished output, securely, in one agent run\u003C/h4\u003E\r\n\u003Cp\u003EThe \u003Cb\u003ECode Execution Tool (in public preview soon)\u003C/b\u003E gives Cortex Agents access to a secure, sandboxed Python environment where they can process data, run calculations, generate visualizations and produce polished outputs — PDFs, PowerPoint presentations, charts — without leaving Snowflake.\u003C/p\u003E\r\n\u003Cp\u003EThe sandbox runs Python 3.12 and includes numpy and pandas by default. Additional packages can be retrieved from PyPI through the Artifact Repository. External network access can be scoped through named integrations. \u003Ci\u003EEach execution runs in an isolated environment. Agent generated code cannot access your data without explicit permissions.\u003C/i\u003E This allows an agent to complete an entire workflow — from analyzing a data set to producing a formatted PDF report — within a single run.\u003C/p\u003E\r\n\u003Ch4\u003EPause, correct and continue your workflow\u003C/h4\u003E\r\n\u003Cp\u003EYou might ask an agent to analyze two years of financial data and then realize you actually need five. Or you notice it heading down an unhelpful path and want to redirect it before it finishes.\u003C/p\u003E\r\n\u003Cp\u003E\u003Cb\u003EInterrupt and Resume (generally available soon)\u003C/b\u003E gives builders a first-class API for human steering. Stop the agent, send a correction, and it picks up from where it left off.\u003C/p\u003E\r\n\u003Cp\u003EThe agent pauses, stores its work and waits for the next instruction, eliminating the need to build custom state management or recovery logic around interrupted workflows.\u003C/p\u003E\r\n\u003Ch4\u003EOne managed agent, multiple access levels\u003C/h4\u003E\r\n\u003Cp\u003EIn most organizations, different users have different permissions. Sales teams may be able to access customer health metrics, while finance teams can also view billing information. Supporting these differences often requires building and maintaining multiple versions of the same agent.\u003C/p\u003E\r\n\u003Cp\u003EWith \u003Cb\u003EPartial Access (in public preview soon)\u003C/b\u003E, the same managed agent can serve requests across different permission levels. When a request arrives, the agent evaluates which tools the user is authorized to use and fulfills the request within that scope. Controls are enforced at the Snowflake layer — consistent with the RBAC already governing the rest of your data. In practice, a single agent can serve both employees and managers, or analysts and executives, from the same knowledge base while enforcing different access boundaries for each.\u003C/p\u003E\r\n\u003Ch2\u003EManage and update your agents\u003C/h2\u003E\r\n\u003Ch4\u003EA visual snapshot of every agent configuration\u003C/h4\u003E\r\n\u003Cp\u003EAgent configurations evolve over time: Prompts get tuned, tools get added, business logic changes. Without structure, teams accumulate undocumented drift between what's running in production and what's being tested.\u003C/p\u003E\r\n\u003Cp\u003EThe \u003Cb\u003EVersioning UI (in public preview)\u003C/b\u003E gives teams a visual history of agent configurations — compare versions side by side, roll back to a previous state and promote a tested candidate to production without editing YAML by hand. A LIVE version stays in production while teams iterate safely on candidate versions before release. For regulated industries, this creates a clear chain of custody for every agent configuration — what changed, when and who approved it.\u003C/p\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/blog/blog-text"},"image_1660099645":{"id":"image-c85adaeec4","height":"692","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--0eef1a1c-06c7-40bc-832e-4d754292e8ff/cortex-agent-versioning.png?preferwebp=true&quality=85","alt":"cortex","width":"1892",":type":"snowflake-site/components/image"},"blog_text_1528741199":{"id":"blog-text-73648e7224","text":"\u003Ch2\u003EBuilt for enterprise scale\u003C/h2\u003E\r\n\u003Cp\u003EThe capabilities we launch today extend the managed platform that Cortex Agents already provides. Orchestration, execution and operational controls help teams scale from their first production deployment to organizationwide adoption.\u003C/p\u003E\r\n\u003Cp\u003EThis has been the main focus for us as Snowflake delivers managed agents: a platform that simplifies the operational complexity, so builders can spend less time maintaining infrastructure and more time building AI applications. As enterprise AI deployments continue to mature, the work shifts from building agents to operating them. That's the next phase Cortex Agents is built to support.\u003C/p\u003E\r\n\u003Ch2\u003EGet started\u003C/h2\u003E\r\n\u003Cul\u003E\r\n\u003Cli\u003ETry the Code Execution Tool \u003Ca rel=\"noopener noreferrer\" target=\"_blank\" href=\"https://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents/code-execution\"\u003Eusing this sandbox configuration and package setup\u003C/a\u003E\u003C/li\u003E\r\n\u003Cli\u003EBuild your first multi-agent workflow \u003Ca rel=\"noopener noreferrer\" target=\"_blank\" href=\"https://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-agents/multi-agent\"\u003Eusing this declarative orchestration guide with working examples\u003C/a\u003E\u003C/li\u003E\r\n\u003Cli\u003E\u003Ca rel=\"noopener noreferrer\" target=\"_blank\" href=\"https://www.snowflake.com/summit/\"\u003EWatch the Snowflake Summit 2026 demos\u003C/a\u003E to see these capabilities running live\u003C/li\u003E\r\n\u003C/ul\u003E\r\n\u003Cp\u003E \u003C/p\u003E\r\n\u003Cp\u003E\u003Ci\u003E\u003Csub\u003EForward-looking statement: This blog post contains forward-looking statements, including statements about products, features and capabilities that are under development, not yet generally available or otherwise not yet available. These forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially. Please refer to Snowflake's SEC filings for a more detailed description of the risks that could cause actual results to differ.\u003C/sub\u003E\u003C/i\u003E\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/blog/blog-text"}},":itemsOrder":["blog_text","image","blog_text_1859485776","image_1660099645","blog_text_1528741199"],":type":"wcm/foundation/components/responsivegrid"},"responsivegrid_premium_content_banner":{"columnClassNames":{},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","appliedCssClassNames":"snowflake-responsive-component-top-padding-medium",":items":{},":itemsOrder":[],":type":"wcm/foundation/components/responsivegrid"},"container_author_chip":{"layout":"RESPONSIVE_GRID","columnClassNames":{"author_chip":"aem-GridColumn aem-GridColumn--default--12"},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","id":"container-c65e9475a1",":items":{"author_chip":{"id":"author-chip-89a4a75387","title":{"id":"title","type":"heading2","lines":["Learn more about the authors"],":type":"snowflake-site/components/title-v2"},"authors":[{"authorImage":{"id":"image-152cab9b6c","height":"653","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--8049a19c-6f69-400d-9f50-c2ee72ba1d88/arun-agarwal.jpg?preferwebp=true&quality=85","alt":"Arun Agarwal","width":"720",":type":"snowflake-site/components/image"},"authorCta":{"id":"button-e3d1624beb","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/arun-agarwal/"},"linkTargetContentType":"DOCUMENT_LEARN",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Arun Agarwal"},"authorTitle":"AI Product Marketing, Snowflake"},{"authorImage":{"id":"image-8e7bdd9fae","height":"400","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--bab3c432-9d00-4fbb-9785-e632873b6c56/harshal.jpg?preferwebp=true&quality=85","alt":"Harshal Pimpalkhute","width":"400",":type":"snowflake-site/components/image"},"authorCta":{"id":"button-0cf52a41a9","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/harshal-pimpalkhute/"},"linkTargetContentType":"DOCUMENT_LEARN",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Harshal Pimpalkhute"},"authorTitle":"Principal Product Manager"},{"authorImage":{"id":"image-b93ca589f5","height":"2549","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--3ab8a4fd-dd09-4fd7-b945-2d43de0f11d8/william-allen.jpg?preferwebp=true&quality=85","width":"2549",":type":"snowflake-site/components/image"},"authorCta":{"id":"button-fa50caa4f2","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/william-allen/"},"linkTargetContentType":"DOCUMENT_LEARN",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"William Allen"},"authorTitle":"Head of Product, AI - Cortex Agents and Snowflake CoWork"},{"authorImage":{"id":"image-7a5c55a5fb","height":"346","lazyEnabled":true,"src":"https://www.snowflake.com/adobe/dynamicmedia/deliver/dm-aid--67ef09e9-67c2-4577-b469-c3cc5c2b457a/vino-duraisamy.png?preferwebp=true&quality=85","width":"346",":type":"snowflake-site/components/image"},"authorCta":{"id":"button-31b2407b6c","showOutboundIcon":false,"buttonLink":{"valid":true,"url":"/en/blog/authors/vino-duraisamy/"},"linkTargetContentType":"DOCUMENT_LEARN",":type":"snowflake-site/components/button","linkType":"SNOWFLAKE_INTERNAL","text":"Vino Duraisamy"},"authorTitle":"Senior Developer Advocate, Product Management at Snowflake"}],":type":"snowflake-site/components/blog/author-chip"}},":itemsOrder":["author_chip"],"appliedCssClassNames":"snowflake-responsive-component-top-padding-medium",":type":"snowflake-site/components/container"}},":itemsOrder":["container_hero","responsivegrid_content","responsivegrid_premium_content_banner","container_author_chip"],":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container"},"flexible_column_content_container_2":{"layout":"SIMPLE","id":"container-e7764747d0",":items":{"blog_table_of_content":{"id":"blog-table-of-content-1732131390",":type":"snowflake-site/components/blog/blog-table-of-content","tableOfContents":[]}},":itemsOrder":["blog_table_of_content"],":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container"},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":true},"related_content":{"id":"related-content-2537410db3","relatedContent":[],":type":"snowflake-site/components/blog/related-content","isBlogPage":true}},":itemsOrder":["flexible_column_container","related_content"],"appliedCssClassNames":"snowflake-container",":type":"snowflake-site/components/container"}},":itemsOrder":["container_breadcrumb","container_main_content"],":type":"wcm/foundation/components/responsivegrid"},"container_47873732":{"additionalClasses":"section--blog-newsletter","layout":"RESPONSIVE_GRID","columnClassNames":{"flexible_column_cont":"aem-GridColumn aem-GridColumn--default--12"},"columnCount":12,"gridClassNames":"aem-Grid aem-Grid--12 aem-Grid--default--12","id":"container-2761da6feb",":items":{"flexible_column_cont":{"id":"flexible-column-container-a2a2ebc140","type":"1-column","alignColumns":"top","containerMaxWidth":"extra-large","topPadding":"small","bottomPadding":"none","spaceBetween":"small","reverseOnMobile":false,"carouselOnMobile":false,"propertiesCSSClasses":"section--blog-newsletter","backgroundImageOption":"none","flexible_column_content_container_1":{"layout":"SIMPLE","id":"container-1443853b63",":items":{"marketo_v2":{"id":"marketo-v2-8b951727cb","marketoForm":{"hidden":null,"formId":"3320","successUrl":null,"edit":false,"script":null,"values":null},"title":{"id":"title","type":"heading3","lines":["Subscribe to our blog newsletter","Get the best, coolest and latest delivered to your inbox each week"],":type":"snowflake-site/components/title-v2"},"munchkinId":"252-RFO-227","serverInstance":"252-RFO-227.mktoweb.com","marketoConfigured":true,"formConfigured":true,":type":"snowflake-site/components/form/marketo-v2"},"text":{"id":"text-c7a6e65860","additionalClasses":"newsletter-disclaimer","text":"\u003Cp\u003EBy submitting this form, I understand Snowflake will process my personal information in accordance with their Privacy Notice.\u003C/p\u003E\r\n","richText":true,":type":"snowflake-site/components/text"}},":itemsOrder":["marketo_v2","text"],":type":"snowflake-site/components/flexible-column-container/flexible-column-content-container"},":type":"snowflake-site/components/flexible-column-container","isActiveTOC":false,"isBlogPage":true}},":itemsOrder":["flexible_column_cont"],"appliedCssClassNames":"snowflake-container",":type":"snowflake-site/components/container"},"experiencefragment-pre-footer":{"id":"experiencefragment-f1ae4a999f","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/get-started-pre-footer/get-started-pre-footer/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/get-started-pre-footer/get-started-pre-footer.xfmodel.json"},"markup_editor":{"id":"markup-editor-4901b047ea","title":"Page CSS","cssContent":"@media screen and (min-width:768px){.snowflake-blog-author-chip-wrapper{justify-content:flex-start}.snowflake-blog-related-content-on-blog-page{max-width:1408px;margin-left:auto;margin-right:auto}.snowflake-text{font-family:Lato,sans-serif;font-weight:400;font-size:16px;line-height:24px}}.section--blog-newsletter{max-width:none;width:100%;padding-left:0;padding-right:0;margin-left:0;margin-right:0;margin-bottom:0}.section--blog-newsletter .mktoField{background-color:transparent !important}.section--blog-newsletter\u003E.container{padding-left:0;padding-right:0}@media screen and (min-width:768px){.section--blog-newsletter\u003E.container{padding-left:0;padding-right:0}}.newsletter-disclaimer p{font-size:14px !important}.section--blog-newsletter .snowflake-marketo-form-container{margin-bottom:24px;background-color:#f6f9fa;gap:48px;box-shadow:none}.section--blog-newsletter .snowflake-title p.snowflake-title-line:first-child{font-family:Texta;font-size:24px;line-height:26px;font-weight:700;margin-bottom:4px}.section--blog-newsletter .snowflake-title p.snowflake-title-line{text-transform:none;font-family:\"Lato\",sans-serif;font-size:16px;line-height:24px;font-weight:normal}@media screen and (min-width:1024px){.section--blog-newsletter .snowflake-marketo-form-container{display:flex;justify-content:center}.section--blog-newsletter .snowflake-title .snowflake-title-line{text-align:left}.section--blog-newsletter .snowflake-marketo-form .mktoFormRow:has(\u003E input[type=\"hidden\"]){flex-grow:0}.section--blog-newsletter .snowflake-marketo-form{display:flex;width:50% !important}.section--blog-newsletter .snowflake-marketo-form .mktoButtonRow{flex-grow:0;width:auto !important;margin-left:0;margin-right:0}.section--blog-newsletter .snowflake-marketo-form .mktoFormRow{flex-grow:1}.section--blog-newsletter\u003E.container{padding-left:0;padding-right:0}.section--blog-newsletter .snowflake-marketo-form-title{width:50%;margin-bottom:0 !important}.section--blog-newsletter .center .snowflake-title{align-items:flex-start}}.snowflake-sub-navigation a.snowflake-sub-navigation-primary-link{width:auto !important}.snowflake-blog-hero{align-items:stretch !important}",":type":"snowflake-site/components/markup-editor","isGSAPEnabled":false},"markup_editor-table":{"id":"markup-editor-47c702bc54","title":"Table Styling CSS","cssContent":"#snowflake-blog-template-main-container table{width:100%;background-color:var(--ui-background-01);border-collapse:collapse;border:2px solid var(--ui-background-09);font-family:'Lato',sans-serif;color:var(--ui-background-09)}#snowflake-blog-template-main-container table thead{background-color:var(--ui-01)}#snowflake-blog-template-main-container table th,#snowflake-blog-template-main-container table td{border:2px solid var(--ui-background-09);padding:var(--spacing-01)}",":type":"snowflake-site/components/markup-editor","isGSAPEnabled":false},"experiencefragment-footer":{"id":"experiencefragment-4e2ed0f3bc","localizedFragmentVariationPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/footer/master/jcr:content","configured":true,":type":"snowflake-site/components/experiencefragment","xfModelPath":"/content/experience-fragments/snowflake-site/language-masters/en/site/footer/master.xfmodel.json"}},":itemsOrder":["experiencefragment-banner","experiencefragment-header","experiencefragment-sub-header","responsivegrid","container_47873732","experiencefragment-pre-footer","markup_editor","markup_editor-table","experiencefragment-footer"],":type":"wcm/foundation/components/responsivegrid"}},":itemsOrder":["root"],":mappedPath":"/en/blog/snowflake-cortex-agents-enterprise-ai-scale/",":type":"snowflake-site/components/structure/page",":hierarchyType":"page",":path":"/content/snowflake-site/global/en/blog/snowflake-cortex-agents-enterprise-ai-scale","analyticsContentTags":["snowflake-site:taxonomy/blog/product-and-technology/ai-ml"],"analyticsEnabled":true,"coveoConfig":{"pipeline":"snowflake.com","apiKey":"xx335921a6-2a0a-40f2-a167-e390b4766c3d","organizationId":"snowflakecomputingproduction8neljofn","searchHub":"snowflake.com"},"analyticsDebugMode":false,"analyticsData":{"excludeFromAnalytics":false,"subCategory":"","pageType":"homepage","templateName":"blog-page","siteName":"snowflake","pageUrl":"/content/snowflake-site/global/en/blog/snowflake-cortex-agents-enterprise-ai-scale","language":"en","category":"general","pageName":"Your Agents Are in Production. Now What?","contentTags":["snowflake-site:taxonomy/blog/product-and-technology/ai-ml"]},"isPasswordProtected":false,"locale":"en"}