MULTI-PROJECT INTERNSHIP
During my internship at B-Robots, I had the privilege of working on diverse projects. Throughout this experience, my primary focus revolved around two significant tasks: data extraction from documents, with a focus on invoices, and working on a computer vision model to integrate within an automated testing framework. For the latter, I collaborated with another intern from a company called Brightest. This experience at B-Robots has proven to be a turning point in my professional journey, providing me with a remarkable platform to enhance and refine my technical skills, expand my knowledge base, and gain invaluable practical expertise in the field. The opportunity to work on such projects alongside a team of dedicated professionals has accelerated my personal and professional growth.
SUMMARY
WHO
B-robots, founded in 2017, has emerged as a prominent player in the field of Robotic Process Automation (RPA). With their deep expertise and innovative approach, they have swiftly established a strong foothold in this dynamic and competitive market. B-robots excels in providing comprehensive solutions across various domains, empowering businesses to streamline their processes and achieve operational efficiency. Their main service, Robotic Process Automation (RPA), enables the automation of repetitive tasks using software robots. This empowers organizations to reduce manual effort, improve accuracy, and enhance productivity. In addition to RPA, B-robots branched out into AI automation, which combines the power of Artificial Intelligence and automation to create intelligent workflows. Leveraging cutting-edge algorithms and machine learning techniques, they enable businesses to automate complex decision-making processes, thereby accelerating digital transformation.
WHY
DOCUMENT DATA EXTRACTION
At its inception, the project was initially envisioned as a Proof of Concept aimed at assessing the feasibility of developing a robust tool to combat Invoice Fraud. B-Robots recognized the critical importance of addressing this pervasive issue, which can significantly impact businesses and their financial well-being. By undertaking this PoC, B-Robots sought to determine whether the creation of an effective solution to detect and mitigate Invoice Fraud was attainable, providing added security and peace of mind to their prospective clients.
As the project progressed, however, the primary goal instead turned to the realms of general Document Data Extraction. The ability to extract valuable information from various types of documents held immense potential for streamlining operations, improving efficiency, and enhancing decision-making processes. The project evolved into a more comprehensive initiative, aligning with B-Robots' vision of developing an in-house tool for Document Data Extraction.
COMPUTER VISION TESTING FRAMEWORK
The process of testing IT applications is of paramount importance as it ensures the reliability, functionality, and quality of the software being developed. However, traditional testing approaches often face challenges in keeping pace with the dynamic nature of modern applications and the ever-evolving technology landscape. In discussions with Brightest, a mutually beneficial opportunity emerged to combine the talents and perspectives of the interns from both companies. The collective goal was to enhance an existing testing framework by integrating an additional capability: the detection of various web elements and leveraging this information to bolster the testing process.
WHAT
DOCUMENT DATA EXTRACTION
The objective of this project has changed throughout its duration. Originally, it aimed to research and address the issue of Invoice Fraud by developing a tool that could detect fraudulent activities on digitalized invoices. By providing this service, B-robots could potentially offer enhanced security measures to its future customers. However, later on the project expanded to encompass general data extraction from various documents, with a focus on evaluating the reliability and validity of extracted information. This would enable B-robots to provide a more comprehensive solution for document processing, benefiting both the company and its customers by streamlining data extraction processes and improving overall efficiency. For this reason, I developed an automated data extraction pipeline using OCR and NLP techniques to extract key information, such as invoice numbers, dates, and amounts, from invoice documents.
COMPUTER VISION TESTING FRAMEWORK
The purpose of this project was to enhance the testing process for web applications by leveraging Computer Vision techniques by working together with another intern at another company. Our goal was to detect and locate clickable web elements within web applications, primarily websites, using purely visual methods. By accurately identifying these elements, we aimed to simulate real-world user interactions by emulating clicks on the identified elements, providing a more authentic testing experience.
EVIDENCE DOCUMENTS
In the subsequent section, you will find downloadable documents that provide comprehensive details about the projects I undertook during my internship. These documents serve as valuable resources, offering in-depth insights into the work I accomplished and the methodologies employed. Feel free to explore and review these materials to gain a deeper understanding of my internship experiences and the projects I contributed to.
