In this step-by-step guide, we'll create a production-ready UiPath data extraction workflow for PDFs and images in just four simple steps. We'll use Cradl AI’s accurate AI OCR (optical character recognition) models to extract key information from any PDF, and seamlessly export the extracted data back into UiPath, letting you automate data extraction with minimal effort.
Software Engineer
UiPath's Document Understanding is a built-in solution for AI-powered data extraction, and enables users to extract data from a variety of common document formats.
Cradl AI is a specialised, standalone SaaS tool designed to seamlessly integrate into any document workflow, offering a best-in-class data extraction experience. It features integrated error handling, a train-as-you-go models that adapt to changing documents, and tools for easy collaboration and monitoring, making it ideal for dynamic and complex document processing needs.
Before we begin, make sure you’ve created a Cradl AI account.
Once you're inside the app, your first step is to create an AI model that understands your documents. In Cradl AI, creating an AI model simply consists of listing the fields you want to extract, and the AI parsing engine will automatically understand what you need.
If you're working with invoices, clone the invoices template. If you're working with purchase orders, clone the purchase order template. You get the idea. Template models offer the quickest way to to great data extraction accuracy due to their prior training on these documents. They're also fully editable.
Many industries operate in environments where document layouts vary a great deal. For these cases, creating a custom model is the way to go. Cradl AI's engine excels at extracting data from such documents and continuously improves over time, learning from your usage to deliver ever increasing accuracy.
In this post, we’re using UiPath Studio Web. If you’re using the desktop version of UiPath Studio, check out this guide.
Once the connection to your Cradl AI model is set up, you’re ready to send your automation.
In addition to automatically uploading documents to Cradl AI, we also want to automatically extract data from them. The Parse document with human-in-the-loop activity let's us do just that, with the added benefit of automatically flagging uncertain predictions for human review, ensuring data integrity.
Once the data has been extracted and validated in Cradl AI, it’s time to send it back to UiPath Studio Web and use it in other automated processes.
To export the extracted data from Cradl AI to UiPath, create a new automation project in UiPath Studio Web. This time, however, we replace the default trigger with Cradl AI's Document Parsing Completed trigger. Use the previously configured Cradl AI connection, and select the AI model you created.
In this example, we print the data in a message box to show how easily Cradl AI output can be used in your workflow. You can write the data to a database, send it to an API, or save it locally. Happy document parsing!
Integrating Cradl AI with UiPath automates document processing by connecting custom AI models to workflows. Create a Cradl AI account, design a model by listing fields or cloning a pre-trained template, and connect it to UiPath via Integration Services. Cradl AI handles data extraction and validation, including human-in-the-loop reviews for uncertain predictions, before exporting parsed data to UiPath. This integration streamlines workflows and offers flexibility for handling diverse documents.
We’ll help get you started with your document automation journey.
Schedule a free demo with our team today!
Internal document processing on autopilot.
