Migrating to microservices The core reason why the platform was not scalable and inefficient was its monolithic architecture. Therefore, our Solution Architect designed and presented a new cloud-native infrastructure of the platform based on Azure Kubernetes, along with the suggested tech stack and the most efficient roadmap. Migrating to microservices allows smooth adding of new SaaS services: anomaly detection, delivery prediction, route recommendations, object detection in logistics, OCR (optical character recognition) of labels on boxes, Natural Language Processing for document verification, data mining, and sensor data processing. DevOps best practices The need for DevOps expertise was identified as another customer pain. Therefore, we are building the DevOps pipeline from scratch, setting up the environment for development and QA in Azure, and introducing CI/CD processes that allow us to easily assemble and deploy microservices to the environment. Computer Vision solution The core component of this project is the Computer Vision (CV) solution for docks that allows contactless tracking of goods with industrial optic sensors and Nvidia Jetson devices. Our client had CV algorithms written by another vendor, which were inefficient and unsuitable for production. Therefore, we found a top-notch CV expert with a Ph.D. degree to run the CV workstream. After careful examination of the existing algorithms, we decided to redevelop them completely. We changed the architecture of the solution and introduced Continuous Delivery for Machine Learning, which allows implementing continuously repeatable cycles of training, testing, deploying, monitoring, and operating the ML models. That is especially important given the global scale at which our client is operating. Multiplatform CV mobile app Also, our team designed the architecture of the multiplatform Computer Vision mobile app and is responsible for its end-to-end development. The app covers object detection, package damage detection, OCR, and NLP for document processing.

The solution is in the development phase. The planned embedded computer vision solution for cameras installed in warehouses will allow our client to automatically detect arriving packages, scan barcodes, and change the delivery statuses of the boxes. And we are currently developing the computer vision pipeline for it. Thanks to the advanced CV algorithms and automated ML system, it is planned that cameras will read the barcodes and textual information under various conditions (i.e. different sizes of the boxes, camera angles, wet or stained labels). At the same time, the multiplatform mobile application will allow warehouse staff to scan barcodes and allocate the boxes efficiently in a warehouse.