A large wholesale distributor runs its core operations on a mainframe, and turning it off means moving every job that mainframe handles onto modern systems first. The company's modernization lead brought us in to do that work, one function at a time. The first project to land came in on schedule with one developer, against the company's own plan for four, and we've now switched off its mainframe processes.
The company's core operations run on a mainframe. Turning it off would save several hundred thousand dollars a year, but only after about seven separate projects move the jobs it handles onto modern systems. And the company didn't have developers to spare for that work.
That's the bind with a system like this. You can't stop the business to replace it, and you can't leave it running forever. The way out was to take it apart one function at a time.
Each project lifts one job off the mainframe, rebuilds it on .NET and Azure, and switches the old code off. So far we've taken on four of them, each with its own scope. The company's modernization lead architects every project, and the company's own team reviews and approves the requirements. We bring the design, development, and testing.
The first project to land. The shelf labels retail locations print now run through a modern API; the company's mobile web app calls it directly instead of writing to a mainframe file, and the old mainframe processes are shut off.
Product, item, and promotion files that feed retail locations' point-of-sale systems, moving off the mainframe onto a new SQL datastore.
Price-change events flow through enterprise Kafka topics into standard, promotional, and personalized price files.
A new numbering system for vendor promotions, with the vendor updates that depend on it.
In the first project, we tested every change before it shipped, with full start-to-finish stress runs. One stress test, a 900-page label report, ran clean.
Almost everything we ship runs on the same modern Microsoft stack: proven, supported, and easy to hire for years from now. Cutting edge but never bleeding edge.
Built inside the company's own environment, on their DevOps, their release pipeline, and their approvals, so each piece goes live where their team already works.
The company's business users have day jobs. On the first project, sprint reviews ran 15 to 20 minutes, business refinement took about 30 minutes every two weeks, and only two technical refinement meetings were needed. Everything else moved over chat. So the company's experts weighed in on every sprint and still had most of the day for their own work.
Tell us the challenge in plain language. We scope the work on complexity, commit to the figure in the room, and stay for the years after launch.
K-Port, the custom EHS compliance portal we have run for an environmental engineering firm since 2016.
Read it →