A warehouse rollout rescued, rebuilt, and shipped to production in weeks.
A fast-growing UK infrastructure operator needed a dispatch-and-returns system live for warehouse staff on the floor. Their Django and React build had stalled under a backlog of defects. We redesigned it, rebuilt it in the toolset their engineers already trusted, and brought it to production, working across every business unit and every integration.
The challenge
Our client, a fast-growing UK infrastructure operator, runs a warehouse that dispatches customer equipment for every new installation and processes every return. The software meant to run that floor was a Django and React application that had stalled. It was carrying a backlog of UAT defects, trust was wearing thin, and a hard deadline to go live was approaching.
We were brought in to land it: not to write a report or advise from the sidelines, but to take ownership and deliver something the warehouse could depend on.
What we did
We decided to redesign rather than patch. Over a few weeks we rebuilt the application as a lean, server-rendered HTMX app, written in the customer's own preferred toolset of Hono and TypeScript, so their engineers could take it on comfortably after handover. There was no framework sprawl and no client-side state to debug late at night: thin routes, business logic ported cleanly into services, and server-rendered fragments swapped in place.
We also worked across every part of the business: not only the warehouse operators, but the customer-service team leads, the business analysts and the client's own engineers. We ran live user testing against real CRM data, tracked every fix as its own ticket on its own branch, and gave the business a clean, auditable record of exactly what changed and why.
- Full redesign and refactor of a failing Django + React system
- Rebuilt in the customer's preferred stack: Hono, TypeScript and server-rendered HTMX
- Three live integrations: a Microsoft Dynamics 365 CRM, carrier shipping labels, and device provisioning
- Disciplined delivery: one ticket and one branch per fix, shared back to the business
- End-to-end collaboration with warehouse, customer service, BA and engineering teams
Built right for production
We didn't just hand over code; we handed over a running production service. We set up CI/CD and deployed on Azure end to end: App Service for the app, Azure PostgreSQL for data, Azure AD for authentication, and a container pipeline through Azure Container Registry. Error tracking and tagged monitoring were in place from the start, alongside an admin console for safely retrying failed operations.
That is why production problems were never a drama. When something did go wrong, as it will in any live system, we knew exactly where to look and could ship a tested fix in around ten minutes. The customer was left with a service that is quick to work on, deployed properly, and that discreetly saves their staff time every day.
This is the service we provide.
End-to-end delivery, in your toolset, seen through to production.
Start a conversation