Summary

At Cuttoshape, I architected a multi-tenant ERP system to centralize fashion business workflows, implementing a layered architecture with strict data isolation and location-aware financial logic. To manage system performance under concurrent request bursts, I built a centralized caching layer using request deduplication and a distributed task orchestration system to offload CPU-intensive processes like document generation and event processing. My work also included engineering a security engine to identify anomalous access patterns and verify media integrity, alongside developing the granular role-based access control required for secure team collaboration and operational delegation.

Highlights

• Architected a multi-tenant ERP platform for fashion businesses, implementing a layered architecture with strict data isolation, location-aware financial logic, and atomic transaction management across complex, high-concurrency business workflows.
• Built a central caching layer using Redis to store frequently read API responses and shared application state. Implemented a promise-based request deduplication system to prevent cache stampedes, ensuring sub-second response times for concurrent requests.
• Engineered a distributed task orchestration layer using BullMQ and Redis to offload CPU-intensive document generation and automated business notifications, maintaining high API availability and throughput during periodic traffic bursts.
• Developed a multi-vector security engine to neutralize platform-wide abuse; utilized device fingerprints and IP heuristics to intercept anomalous account-level patterns, alongside a multi-stage media pipeline featuring pre-upload validation and post-upload integrity verification.
• Implemented a secure staff management system that allows business owners to delegate operational tasks through granular role-based access control and managed invitation workflows, enabling secure collaboration for growing fashion teams.
• Implemented a comprehensive testing strategy consisting of isolated unit, cross-module integration, and high-concurrency stress tests to ensure system reliability, performance stability, and data consistency across all application modules.