{"id":24852,"date":"2026-06-22T14:49:03","date_gmt":"2026-06-22T12:49:03","guid":{"rendered":"https:\/\/www.atamya.com\/?page_id=24852"},"modified":"2026-06-26T15:30:17","modified_gmt":"2026-06-26T13:30:17","slug":"ai-powered-workflows","status":"publish","type":"page","link":"https:\/\/www.atamya.com\/en\/agentic-pim\/ai-powered-workflows\/","title":{"rendered":"AI-Powered Workflows"},"content":{"rendered":"\n
\n
\n
\n
\n

AI-Powered Workflows<\/h1>\n

AI as a process. Not as a button.<\/strong>
\nIn many PIM systems, AI lives in a separate window. In ATAMYA, however, it operates directly from within the process itself \u2013 as a native service task, orchestrated by a full-fledged BPMN engine. Sounds like an infrastructure detail. Indeed, it is. And that\u2019s why it works.<\/p>\n

See Our AI Workflow Live<\/a><\/p>\n\n <\/article>\n

\n
\n \"\"\n <\/figure>\n <\/div>\n <\/div>\n <\/div>\n<\/section>\n\n\n
\n
\n
\n
\n

AI Service Task: No Prompt Window, but a Process Step.<\/h2>\n

Imagine a classic AI chatbot in a PIM. Prompt comes in, text comes out, copy, then paste. While this might indeed be helpful, it is far from an integrated process: No system. No context. No traceability.<\/p>\n

An AI service task in ATAMYA functions differently. It\u2019s a BPMN service task \u2013 just like “Active Object,” “Edit Attribute Value,” or “Synch with Shopware.” It has inputs, outputs, as well as conditions before and after. You can run it parallelly, embedded it in loops, or escalate it in error cases. It writes values as variables that can be referenced in subsequent workflow steps.<\/p>\n

In short:<\/strong> AI becomes an integral component of the workflow engine and takes care of your complete orchestration logic.<\/p>\n\n <\/div>\n <\/div>\n <\/div>\n<\/section>\n\n\n

\n
\n
\n
\n

What AI is Already Accomplishing Today in ATAMYA<\/p>\n

ATAMYA offers multiple AI-native tasks \u2013 fully configurable components with prompts, variables, targeted outputs, and guardrails:<\/p>\n

    \n
  • Generate Text Content:<\/strong> Free-text prompt with dynamic variables, the result is assigned directly to a variable.<\/li>\n
  • Translate Texts:<\/strong> Dedicated translation tasks, optionally with DeepL for maximum quality, including your business terminology.<\/li>\n
  • Extract Attribute Values from PDFs:<\/strong> Product datasheets or supplier documents can be read automatically \u2013 the AI identifies attributes and adds their values to relevant objects.<\/li>\n
  • Decision-Making Aid:<\/strong> AI can answer only with true or false \u2013 directly useable in the following gateway.<\/li>\n<\/ul>\n

    Every task can be repeated as many times as necessary, connected to conditions, or supplied with live data from your PIM via our expression language.<\/p>\n\n <\/article>\n

    \n
    \n \"\"\n <\/figure>\n <\/div>\n <\/div>\n <\/div>\n<\/section>\n\n\n
    \n
    \n
    \n
    \n

    When AI Cooperates in Process Decisions<\/h3>\n

    In ATAMYA, AI does not only generate content but prepares decisions directly in your workflow. AI responses land in variables and flow into gateway conditions via our expression language.<\/p>\n

    An example: As soon as a new product is created, a first AI task generates the product description. A second one validates the regulatory requirements and responds with true or false.<\/p>\n

    The gateway decides automatically:<\/p>\n

      \n
    • true<\/strong> \u2192 object is activated automatically<\/li>\n
    • false<\/strong> \u2192 a user task is sent to the technical writers, including missing aspects identified by a third AI call<\/li>\n<\/ul>\n

      Three service tasks. One gateway. Not a single line of custom code. And the editing team only receives cases that truly require their attention.<\/p>\n\n <\/article>\n

      \n
      \n \"\"\n <\/figure>\n <\/div>\n <\/div>\n <\/div>\n<\/section>\n\n\n
      \n
      \n
      \n
      \n

      Expression Language: AI Straight from Your PIM<\/h3>\n

      What makes an AI task in ATAMYA so powerful is not simply that it is AI-based but the expression language upon which it is grounded. It accesses your product data and renders prompts dynamic: no static texts but an intelligent combination of data, contexts, and rules.<\/p>\n

      Here is what else you can map with expressions:<\/p>\n

        \n
      • Dynamic attribute values<\/strong> that can be used in prompts \u2013 including data languages and contextualized attribute values<\/li>\n
      • Set and read variables<\/strong> \u2013 to use them in subsequent workflow steps<\/li>\n
      • Computing numeric values<\/strong> \u2013 DQM scores, degree of completion, or confidence rating<\/li>\n
      • Combination of multiple attributes, objects, and domains <\/strong>in a single call<\/li>\n<\/ul>\n

        The result: One-time set up, applicable dynamically to thousands of cases. For each product, each domain, and each language.<\/p>\n\n <\/article>\n

        \n
        \n \"\"\n <\/figure>\n <\/div>\n <\/div>\n <\/div>\n<\/section>\n\n\n
        \n
        \n
        \n
        \n

        Human-in-the-Loop is Part of the Process, not the Exception<\/h3>\n

        \u201cAI is great \u2013 but only if I can double-check its results.\u201d<\/em> A justifiable approach. This is why human-in-the-loop is not a safety net in ATAMYA but, much rather, a native workflow step.<\/p>\n

        Every AI output can be directly written into a user task. People in response can view the result to either accept, adjust, or decline it \u2013 the decision then flows into a workflow variable, and the process continues.<\/p>\n

        Practical example: The AI suggests a classification. If the confidence score is high enough, it will be processed automatically. If it is below the user-defined threshold, a review is generated for the product manager \u2013 with AI suggestions as the starting point, not the decision itself.<\/p>\n

        People are not replaced. They receive the tasks that actually matter.<\/strong><\/p>\n\n <\/article>\n

        \n
        \n \"\"\n <\/figure>\n <\/div>\n <\/div>\n <\/div>\n<\/section>\n\n\n
        \n
        \n
        \n
        \n

        Four Paths to Start a Workflow<\/h2>\n

        AI steps live in workflows \u2013 and workflows can be started in four different ways in ATAMYA.
        \nBe it by a person with a manual mouse click, a user-defined schedule, or an external system call \u2013 the workflow engine listens for the event trigger, fills out the start variables, and boots the process. With or without AI tasks. The orchestration remains one and the same.<\/p>\n

         <\/p>\n\n <\/div>\n <\/div>\n <\/div>\n<\/section>\n\n\n

        \n
        \n
        \n
        \n \n
        \n \"\"\n

        Manually<\/p>\n

        via a button on the relevant business object\u2019s or group\u2019s details page<\/p>\n<\/figcaption>\n <\/figure>\n <\/div>\n

        \n \n
        \n \"\"\n

        Time-Controlled<\/p>\n

        via timer definitions, scheduling, say, a DQM check overnight at 3:00 AM<\/p>\n<\/figcaption>\n <\/figure>\n <\/div>\n

        \n \n
        \n \"\"\n

        From Inside<\/p>\n

        via signal start events triggered upon a user-defined event like product creation, attribute change, or state changes<\/p>\n<\/figcaption>\n <\/figure>\n <\/div>\n

        \n \n
        \n \"\"\n

        From Outside<\/p>\n

        via REST calls by external systems or our MCP server<\/p>\n<\/figcaption>\n <\/figure>\n <\/div>\n <\/div>\n <\/div>\n <\/section>