How Artificial Intelligence is Transforming Automation in IT
Documentation in Automic has always been a challenge, and AI is now changing the way it can be managed. Artificial Intelligence began to reshape the IT landscape in the early 2020s, and its influence has only grown stronger since then. One of the areas where this transformation is most visible is automation.
From Automation to Intelligent Operations
The integration of AI into the Automation Engine (AE) opens up new possibilities. It demonstrates that automation is no longer just about executing processes, but also about generating value through deeper insights and intelligent support. One of the areas where this becomes especially impactful is documentation. Clear and structured documentation has always been critical for IT infrastructures, ensuring that jobs and workflows remain maintainable over time, even if the original creator is no longer available.
The Documentation Challenge
Despite its importance, documentation is often overlooked or minimized. Developers frequently skip it in order to save time, but this short-term gain can create long-term risks, such as knowledge gaps, slower onboarding of new team members, and higher maintenance costs. This is exactly where GenAI can provide a breakthrough — by making documentation faster, easier, and in many cases even automatic.
How GenAI Enhances AE Workflows
A well-structured AE workflow enhanced with GenAI can automatically generate documentation as part of the job execution process. This ensures that important details are never missed and that accurate, up-to-date information is always available.
Step 1: Activating the Job Plan
The process begins with the activation of the job plan. As soon as this step is triggered, the workflow prepares the necessary information for documentation, making it a natural part of the process rather than an afterthought.
Step 2: Defining the Object for Documentation
Once activated, the object to be documented must be defined. This is achieved through a combination of a PromptSet and an SQLI VARA. The SQLI VARA runs a query to gather potential objects for documentation, which are then presented in a combo box for easy selection. At this point, specific objects can also be excluded from the dialog if they are not relevant for documentation.
From Automation to Intelligence
This approach illustrates how AI is not only streamlining workflows but also redefining the role of IT professionals. By embedding GenAI into AE, documentation becomes a reliable, automated output that removes one of the most common weaknesses in IT operations. The result is greater consistency, reduced risk, and more time for IT teams to focus on innovation instead of repetitive work. In this way, AI-powered automation helps organizations adapt more quickly to change while building a stronger foundation for long-term success.
Once the setup is complete, configure the PromptSet with the correct DataReference—the name of the SQLI VARA—and the Variable Name, which will later be referenced in the job plan. After saving these settings, the automation workflow can begin.
The first action is to export the script content of the selected object. For this, AE’s built-in EXPORT() procedure is used, which accepts the object name from the PromptSet variable and a destination file where the system writes the extracted script.
Once the content is available, it must be parsed. In this case, a simple self-built Python parser with the xml.etree.ElementTree module was used, though any XML parsing solution would work — at the end of the day, it just has to function reliably. The parser stores the script in an external file, and another job reads this file and loads the content into AE variables so that it can be processed by the AI framework.
If the script content is not empty, the AI is called via the ASK_AI() procedure. Here, the AI is instructed to generate documentation for the extracted script. Since AE has limitations, the response is restricted to 1024 characters to avoid truncation. The AI is also asked to avoid characters that might break the XML structure. Once the documentation is generated, it is written back into the XML file under the correct documentation section using the same Python parser as before. The final step is to re-import the object into AE.
For this, the built-in IMPORT() procedure is used with the path to the updated XML file. The folder argument can be left empty, but the Object Setting must be set to 1, otherwise the import will skip the object since it already exists. The result of this job plan will appear in the documentation pane of the selected object.
The result shown above is generated from the following script contents:
Fine-tuning AI for Structured Documentation in Automic
This solution can be fine-tuned to create structured documentation, or enriched with predefined information such as priority, severity, or responsible person. This allows teams to create consistent and detailed documentation that goes beyond technical details and integrates operational relevance.
Recommendations for Reliable Usage
Although this approach significantly speeds up the documentation phase of development, it is always recommended to carefully review the results of the job plan. Documenting multiple objects at the same time is not advisable, as it increases the risk of inaccuracies and reduces transparency.
Beyond Documentation: Broader Benefits of AI in AE
Speeding up documentation is just one area where we can enjoy the benefits of AI in AE. There are several additional opportunities hidden behind it that can deliver even greater value.
AI-Driven Alerting
Built-in AI analysis attached to the alerting system can be extremely useful. Operators and administrators already have a bunch of information at hand when an alert arrives — including error descriptions, possible root cause analyses, and suggested solutions. This drastically reduces troubleshooting time and enables faster resolution.
Custom Models with Ollama
With Ollama integration, companies can train their own custom models and seamlessly use them within Automic. This means AI in AE can be tailored to specific organizational needs, creating more relevant and context-aware automation support.
Expanding Use Cases
AI in AE will not only support error analysis and software development but can also be applied to broader business areas such as data analysis, data mining, trend forecasting, and even automated decision-making. This unlocks a much wider range of possibilities for enterprises looking to innovate.
Conclusion
As AI becomes more intelligent, its combination with Automic opens new opportunities for smarter, faster, and more reliable processes. Organizations that position themselves as pioneers in AI adoption stand to gain a substantial competitive edge by unlocking major efficiencies, saving time, and reducing resource consumption.
Want us to help you explore advanced AI functions in Automic? Let’s talk.
Author: Ádám Murányi
is a technology professional with expertise in automation and AI integration, helping organizations streamline workflows and improve documentation quality.
