A provider of freight brokerage services
Logistics & Transportation
Software Development
Google Cloud, Python, RestAPI, JMagick, SageMaker


Our client — a US-based provider of logistics services — was at a breaking point. The influx of documents from thousands of shippers and carriers was overwhelming their outdated, inefficient document management system. Manual processes used for cumbersome and time-consuming tasks like document indexing were holding them back. As the number of documents continued to rise, the urgency to switch to automated technology only grew stronger. The company turned to ITRex for ML development and computer vision services to build a proprietary OCR document management solution that would automate their processes and improve productivity.

Our task was to:
Conduct an in-depth examination of the client’s diverse data sources to identify and prioritize document types for incremental delivery to ensure a focused and effective rollout
Assess the variety of unique client formats to tailor the OCR solution
Develop bespoke ML algorithms to extract the necessary info for the Client from a recognized document
Train and fine-tune the ML models to handle diverse document types and formats with high accuracy
Design an intuitive and user-friendly interface for OCR tool management and monitoring, accessible to non-technical users


ITRex has developed a cutting-edge OCR document management solution, designed to seamlessly integrate into the client’s existing internal systems. This proprietary solution stands out with its user-friendly interface, accessible both to tech-savvy and non-technical users, and its ability to recognize an extensive range of document and data types with remarkable accuracy. Key features and capabilities: Versatile document recognition:
Capable of processing nearly 20 document types from carriers and shippers, with some types generated by up to 6,000 unique customers using unique formats
Achieves up to 90% recognition accuracy, enhanced through comprehensive training on a diverse dataset
Advanced ML for classification:
Classifies document types and identifiers, with the ML model trained on hundreds of documents to scale to over a million documents
Efficient processing:
Prioritizes documents for recognition based on length, optimizing processing time
Implements distributed processing for increased fault tolerance and scalability to ensure no loss of documents in the recognition pipeline, even as processing demand increases
Includes document rotation for enhanced recognition efficiency
Recognizes every field within a document, a significant improvement from the previous capability of recognizing only four fields manually
Identifies document quality (poor-quality documents, e.g., damaged or torn)
Generates dates and numbers in unified formats through post-processing
Utilizes ML models to recognize handwritten text and signatures
Integration with communication platforms:
Seamlessly integrates with messaging systems like Slack for real-time performance reporting
UI design:
Easy to navigate for non-tech users
Advanced features for more experienced users, allowing manual labeling (tagging) of unique document templates
Includes a dedicated monitoring page to track the performance and efficiency of the AI/ML engine
transportation software development
logistics software development


A dramatic reduction in operational costs through automation of manual processes that not only saves time but also minimizes the likelihood of human error
Enhanced productivity with a greater speed and accuracy of document processing, leading to a more streamlined workflow
The solution has laid the foundations for implementing a comprehensive big data and reporting platform for deeper insights into operational efficiences

Latest projects