Metalex: Industrial SaaS Platform
The Bottleneck
"Small to medium metal manufacturers often rely on paper-based tracking for raw materials and scrap, leading to massive data inaccuracies. For Metalex, processing a single invoice took up to 48 hours because it required manual verification of production logs across multiple departments. The lack of a real-time production timeline resulted in frequent delivery delays and frustrated B2B clients."
Key Features
Automated Metal Scrap & Production Tracking Pipeline
Batch Invoice Processing with Dynamic Tax Engine
Real-time Production Timeline & Workflow Visualization
Multi-tenant Architecture with Branch-level Isolation
Comprehensive Exportable Data Insights for Financial Audits
Digital Weighbridge Integration for Automated Data Entry
The Architecture
I designed a robust SaaS platform using TypeScript and Prisma ORM to provide complete type-safety across the stack. I implemented a multi-tenant architecture that allows the platform to serve multiple branches with strict data isolation. I engineered a 'Batch Processing' engine that automates tax calculations and invoice generation for hundreds of production units simultaneously. To solve the timeline problem, I built a custom Gantt-style production tracker that visualizes the status of every metal batch from 'Melt' to 'Shipping'. The backend utilizes PostgreSQL with advanced indexing to handle complex aggregations of metal weights and scrap percentages in real-time.
Overcoming Challenges
Type Safety: Managing complex unit conversions (MT, KG, PCs) across the entire application using a unified TypeScript system.
Multi-tenancy: Implementing efficient data filtering at the Prisma middleware level to prevent cross-tenant data leaks.
Query Optimization: Fine-tuning recursive SQL queries to calculate production yields across multiple stages of manufacturing.
Protocol & Process
Phase 01: Process discovery at the factory floor to understand the physical production flow.
Phase 02: Developing the multi-tenant core and authentication infrastructure.
Phase 03: Building the automated invoicing and industrial calculation engine.
Phase 04: Frontend development of the production visualization tools.
Phase 05: Deployment on AWS with automated backup and disaster recovery protocols.
Engineered Impact
Document processing time was reduced from 48 hours to just 15 minutes. The production team now has 100% visibility into material loss, allowing them to optimize the melting process and reduce scrap by 12%. The platform has become the backbone of the factory's digital transformation, enabling them to scale their production volume by 30% without hiring additional administrative staff.