
WMS360
UK cloud-native SaaS — real-time inventory & order operations across 11+ global marketplaces. Zero oversell guarantee, sub-100ms API response.
Overview
WMS360 is a cloud-native warehouse and multichannel inventory management SaaS built for UK sellers running operations across a dozen different marketplaces at once. The core problem it solves is deceptively hard: keep stock levels, orders, and fulfillment perfectly in sync across eBay, Amazon, Shopify, TikTok Shop, OnBuy, Wowcher, GoGroopie, and Fruugo — simultaneously, in real time, without ever overselling a single unit.
I joined as the engineer responsible for the inventory sync engine, order pipeline, and warehouse operations tooling, working directly with a UK-based product team in a remote Agile/Scrum setup.
Key Features
- Real-time inventory synchronization across 11+ marketplaces, processing hundreds of thousands of stock update events daily with a zero-oversell guarantee under high concurrency.
- Event-driven order pipeline built on RabbitMQ — guaranteed delivery, dead-letter queues, and automated retry logic to absorb burst order volumes without dropping orders.
- Unified order management with batch processing, invoice generation, packing slip printing, and automated dispatch workflows spanning every connected channel.
- Multi-location warehouse operations — barcode scanning, pick-list generation, bin/shelf mapping, stock transfers between locations, and dynamic low-stock alerting.
- Multi-carrier shipping integration across 10+ UK couriers with real-time tracking, thermal label printing, and webhook-based tracking ingestion.
- SKU-level analytics — channel performance breakdowns, revenue dashboards, demand forecasting, and automated daily reporting exportable as CSV/PDF.
- AI-powered customer service module using the OpenAI API for automated query responses and support ticket classification, meaningfully cutting manual support load.
Technical Highlights
- Concurrency-safe stock reservation logic to guarantee zero oversell even when the same SKU sells across multiple channels within milliseconds of each other.
- RabbitMQ producers/consumers with dead-letter queues and exponential-backoff retry strategies for resilient background processing.
- Redis caching and connection pooling to hold API response times under 100ms even under concurrent load.
- MongoDB and PostgreSQL used side by side — document storage for flexible marketplace payloads, relational storage for transactional order and financial data.
Impact
The platform now underpins day-to-day operations for UK multichannel sellers, replacing manual spreadsheet-driven stock reconciliation with an automated system that keeps inventory accurate to the unit across every channel — the single most common (and costly) failure mode in multichannel e-commerce.


