Maia, Matillion's powerful AI assistant for building, fixing, and managing pipelines with natural language, becomes even more effective enhanced with context files. Context files allow Maia to apply business rules and significantly reduce production times by embedding reusable knowledge and best practices directly while also enabling the override of Maia's behavior with domain specific knowledge and more specific reusable instructions. Context files in Maia are reusable markdown files stored in your project that provide guidance like data standards or naming conventions. Maia reads these files automatically and uses them to tailor its responses. This way you don't have to repeat the same requirements in every prompt. Using context files helps you follow organisational best practices avoid repeating instructions, therefore decreasing even more production time using Maia. In this video I'll demonstrate how to use context files to build a pipeline that downloads data from s three into a table. The data includes information on previous deliveries that were delayed and the weather conditions at the time. This table will later support the creation of a delay probability table using Maia. While the final goal is to use Maia to help build that delay probability table, the focus of this tutorial is specifically on context files and how they can enhance our workflow with Maia. Using context files gives Maia relevant information upfront about how you want your pipelines configured. This helps automate more of the pipeline setup Process. Reducing the number of manual changes you need to make Without a context file, you might find yourself repeating the same configurations like specifying the same storage URL across multiple Excel query components, which Maia could have handled automatically. In short, context files reduce manual work by letting Maia know your preferences ahead of time. So to start, let's talk about how you order and manage your context files. You can do this from the files panel by navigating to the Matillion forward slash Maia forward slash rules directory in your project. If you don't see this folder right away, you can create it manually or ask Maia to do this for you as I will do now. In my case, I'm also going to ask Maia to create multiple context files for different purposes. However it is possible to have just one context file if you want to. I'm going to have context files for naming connections and standards. By doing this, I make it easier to update and find key information specific to that context file. Every time Maia runs, it will automatically read all .md files in the rules folder and apply them to every prompt. So in my case, I want to make sure that specific settings are configured automatically when a certain component is added to the canvas by Maia. The component I want to do this for is the Excel query component. And I want to set the storage URL to the S3 bucket location that contains files relevant to my project. So I will simply make an Excel query component settings section within my connections context file and specify this. Now if I ask Maia to create an Excel query component for me, it will always set the URL to what I specified as we can see here. Our context files could be filled with information on lots of components, us to further streamline our pipeline creation process. They could also include information on the wider project and its objectives, as well as how you want Maia to interact and speak with you. A variety of context file examples and helpful information are available on the Matillion Exchange, which is accessible through the link provided below this video. Now that the Excel query component is added to the canvas, let's take a look at it. As we can see, the component has been filled in with the settings provided in the context file. However, Maia has also intelligently filled in the rest of the properties for me, including the data source, data selection and target table properties. I'm happy with the properties it has filled in for me so I will validate the pipeline and sample the data of the component. I can see that the data is being loaded successfully, so I will run the pipeline. As we can see from sampling the loaded data, The Excel file that has been loaded from S3 has been successfully loaded into the table. Context files can be used to specify much more information than the example we have just used. You can equip Maia with a domain specific understanding of user roles, data sources, tools and workflows. Allowing it to respond appropriately to natural language input, trigger automation and deliver situationally aware insights. In the context file you can see here, we are providing Maia with structured background knowledge about a specific user role, which in this case is a data analyst. We are also contextualizing the data environment and operational workflow. There are multiple examples of context files on the exchange which are available to download. Additional examples include integrating Matillion's data preparation capabilities with AI driven automation for a data scientist and detecting and remediating data quality issues for data quality engineers. So now you know how to create context files. You can begin streamlining your own workflows with Maia. If you want to learn more about Maia and its capabilities, go to www.maia.ai/?utm_source=Matillion&utm_medium=referral&utm_campaign=migration26