In this tutorial, we'll be using Maia, our agentic AI, to build an intelligent orchestration pipeline. This time, we are enhancing the pipeline with additional data on sales volume and annual customer spend, which will later help us analyze delayed deliveries in relation to larger orders. We'll give Maia enough context to complete this task for us. We will let Maia know where the data lives and Maia will handle the rest. As a reminder, Maia takes a highly autonomous approach to building and editing data pipelines, and responses may vary slightly from run to run. This tutorial shows just one example of how Maia might interpret and act on this request. In our scenario, we're continuing work for Greenwave, a national e-commerce retailer working to improve customer satisfaction by proactively addressing delivery delays. To do that effectively, GreenWave wants to bring in richer order context. Specifically, how much each customer is spending and how often they're ordering. The data we are retrieving contains information on how much our clients have spent on our products each quarter. Let's start by opening up our Excel data load orchestration pipeline. You will remember that we used this pipeline to gather our data on shipment weather delays. Our data on the sales volume and annual customer spend of each of the GreenWave customers is also in S3 in an Excel file. So in this case, we are going to ask Maia to extract that data for us. We are also going to specify that we want this to be built in a new orchestration pipeline. As you can see, Maia has got to work straight away, and after a short amount of time, Maia has successfully integrated the Excel Query component. If we dive into the properties of the component, we can see that Maia has intelligently configured some of the properties for us. For instance, in the Configure expandable section, we can see that data source has been selected. The data selection property has also been configured to return all the columns from the query. The database and schema properties in the destination expandable section have also been set for me. In my case, these have been set correctly. Finally, Maia has also given us an appropriate name for the target table. If we sample the data from the component, we can see our data is being loaded successfully. Now if we run the pipeline, we can see in the task history that it ran successfully. Also, if I open the schemas panel, I can see that the table is present in my schemas folder. If we create a transformation pipeline and use a table input component, we can see the new table has now been loaded into our data warehouse by sampling the component. So that covers how to further develop a pipeline using Maia and only essential information. If you want to learn more about Maia and its capabilities, go to matillion.com/maia.