Igor Ferreira Pinto

The project consisted of receiving a high volume of PDF files with new customers' information to register them as new customers for the current company.

The RPA solution was designed to scrape the PDF file for personal information and then register each customer through an internal web portal and track the registration status. The project yielded over 1,000% revenue over the year for the referred process and was elected the best improvement for 2021.

The robot extracted from SAP reports with the stock of all products in multiple warehouses and factories.

Based on these reports, the robot calculated and requested the appropriate redistribution of products through the web portal. This project involved a high volume of items and overly complex business rules.

It is essential that the business has a clear view of the work done by robots and how effective the robots are operating.

For this reason, a robot was created to extract historical data of RPA executions and build an interactive dashboard with a business view of the processes.

In a second phase, this dashboard was integrated with PowerBI, having an even more powerful and dynamic interaction for the users.

Designed and delivered an enterprise automation solution to streamline invoice intake within a global Procure-to-Pay operation. The solution processed invoices received through email as PDF attachments and ZIP archives, extracting relevant invoice information and supporting downstream validation and matching against SAP ECC records.

Key contributions:
• Automated document intake and processing across multiple email attachment formats.
• Supported extraction of key invoice fields, including vendor, invoice date, amount, and purchase order references.
• Integrated automation workflows with SAP-based validation and matching processes.
• Helped reduce manual handling and streamline invoice processing for accounts payable operations.

Contributed to a proof of concept combining Automation Anywhere and AI-powered document processing to reconcile contract information extracted from documents against metadata maintained in SAP Ariba.

Key contributions:
• Applied LLM-based document extraction to identify relevant contract information.
• Automated reconciliation between extracted contract data and existing Ariba metadata.
• Incorporated human validation into the workflow to review discrepancies and support reliable outcomes.
• Explored the application of AI-assisted automation to improve contract data quality and reduce manual reconciliation effort.