Warehouse management, Industrial Manufacturing, Supply Chain, Logistics, Web Development
Angular (SPA), TypeScript, NestJS, PostgreSQL, TypeORM, Firebase, OpenAI GPT API (for OCR)
A large industrial equipment manufacturer operating several warehouses across different regions. The client required a modern, unified system to replace a legacy inventory process that relied heavily on manual data entry, lacked real-time visibility, and offered no mobile support for warehouse staff.
The business needed a reliable and scalable web platform to support real-time stock visibility, accelerate warehouse workflows, and reduce human error.
The client faced several operational issues:
To address the client’s operational challenges, we designed and developed a full-featured web application tailored for warehouse inventory and container tracking. The architecture was built using Angular on the frontend and NestJS on the backend to ensure both performance and scalability across multiple warehouse locations.
The frontend application is a single-page interface (SPA) built with TypeScript and Angular, offering a fast, interactive, and user-friendly experience. It was optimized for use on mobile devices and tablets, enabling warehouse staff to work directly from shop floors, loading docks, and storage areas without the need for desktop access. Features like responsive design, real-time data rendering, and barcode scanning support ensured that users could perform check-ins, order picking, and stock updates quickly and with minimal friction.
The backend, powered by NestJS, handles all business logic and data communication. It connects to a PostgreSQL database using TypeORM, enabling relational data management with a strong schema and easy scalability. PostgreSQL was chosen for its reliability, extensive community support, and rich querying capabilities, which are crucial for handling complex inventory operations across distributed warehouses.
Firebase was integrated to handle authentication. It supports both email/password login and Google account authentication, simplifying access for users and making the system more secure. The system is also designed to support additional OAuth providers (like Apple ID and Amazon) if needed in future expansions.
A major innovation in the solution is the use of OpenAI’s GPT API to perform OCR (Optical Character Recognition) on scanned delivery notes and shipping documents. Instead of building a custom-trained OCR pipeline from scratch, we leveraged GPT’s language understanding capabilities to extract and structure key inventory information directly from images and PDFs. This significantly reduced development time and delivered high accuracy in real-world conditions.
Real-time updates were made possible through Firebase’s realtime database features, allowing all users to see inventory changes, order status updates, and picking lists immediately after any action is taken. This eliminated delays between actions in the warehouse and their visibility to management or logistics coordinators.
The implementation of the new system significantly streamlined the client’s warehouse operations, cutting down manual labor and virtually eliminating spreadsheet-based tracking. Staff were able to perform check-ins, update inventory, and fulfill orders directly from mobile devices, reducing human error and improving task efficiency across all warehouse locations.
Real-time inventory synchronization and AI-powered OCR drastically reduced stock discrepancies and data entry mistakes. Automated document processing and digital workflows enabled faster order picking and shipping, resulting in quicker dispatches and improved customer satisfaction.
In addition to immediate efficiency gains, the platform provided leadership with full operational visibility through live dashboards and centralized data. This transparency supported better decision-making and scalability, laying a strong foundation for further automation and digital transformation.