Matthews Automation

Background

Matthews Automation Solutions, a subsidiary of Matthews International, manufactures industrial thermal inkjet (TIJ) printers used on production lines worldwide. Their Redimark and Evolution brands print expiration dates, lot codes, and barcodes on everything from beverage bottles to pharmaceutical packaging.

The company had a problem they couldn't solve internally: global teams across multiple campuses had built divergent systems with incompatible architectures. Software, mechanical, and electrical engineering operated in silos. Quality assurance meant manual regression testing on physical hardware. And a legacy UI stack made it nearly impossible to hire front-end developers.

They needed someone who could work across all of it: embedded Linux, industrial UI, hardware integration, and team dynamics. And they needed to do it without disrupting live production.

The situation when we arrived

Matthews had talented engineers, but years of distributed development had created fractures:

Siloed global teams meant each campus had its own architecture and processes. Changes in one location broke things in another. Nobody owned the whole picture.

Manual testing on hardware turned every release into a multi-week event. Engineers would physically connect printers, run tests by hand, and hope nothing regressed. Bugs slipped through anyway.

A legacy UI stack that predated modern web standards. Finding developers who wanted to work in it was hard. Training them took months. Turnover was high.

Brand inconsistency across Redimark and Evolution interfaces. Same company, same users, different visual languages. Operators had to relearn controls when switching between product lines.

What we did

This wasn't a six-month engagement with a tidy handoff. We embedded with Matthews for over a year, working across seven distinct workstreams simultaneously.

Unified the teams around two-week demos

We introduced an agile cadence across software, mechanical, and electrical engineering. This is uncommon in hardware organizations, where waterfall still dominates. Every two weeks, a working prototype. Not a slide deck. Not a status report. A prototype you could touch.

This forced integration early and often. Problems surfaced in demos, not in production.

Built architecture that let campuses collaborate

We brokered a system design that separated the UI from deeper control systems. Each campus could move independently without breaking each other's work. The key: designing for win-wins, not winner-take-all. No team felt steamrolled.

Automated testing on the actual hardware

We connected production printers to a CI server. Regression suites now run directly on the devices, automatically, on every commit. This is rare. Most embedded teams treat automated device testing as too hard to implement. We didn't.

Quality went from a periodic event to an everyday habit.

Replaced the legacy UI with web technologies

We moved the interface layer to Chromium with HTML, CSS, and JavaScript. The immediate benefit: Matthews could now hire from a vastly larger talent pool. The longer-term benefit: faster iteration, better maintainability, and a foundation that won't need replacing in five years.

Designed for operators in gloves

Industrial UI has constraints consumer apps don't. Operators wear gloves. They're 20 feet from the screen. They're checking the line, not staring at a dashboard. We redesigned both Redimark and Evolution interfaces around high-visibility gauges, large touch targets, and at-a-glance status indicators.

The result: a cohesive design system across both brands. Same company, consistent experience.

Hardened the operating system

We rebranded an Ubuntu-based distribution and configured it in kiosk mode. This eliminated attack surfaces, simplified updates, and reduced the operational burden on Matthews' IT team. Devices now boot directly into the application with no desktop access.

Recruited and mentored their team

We didn't just build. We hired. We brought in senior engineering leadership and recruited full-time contributors to close skill gaps. Then we mentored internal staff so Matthews could sustain momentum after we stepped back.

Results

Faster delivery. The two-week demo cadence created accountability and surfaced integration problems early.

Higher confidence in releases. Automated on-device testing catches regressions before they ship.

Easier hiring. A mainstream web stack means Matthews competes for the same talent as every other tech company, not a shrinking pool of legacy specialists.

Unified brand experience. Redimark and Evolution now share a visual language and interaction model.

Simpler operations. The kiosked Ubuntu distribution reduced IT burden and security exposure.

Matthews emerged with a modernized platform and teams equipped to extend it, without sacrificing the rigor required for mission-critical industrial hardware.

What made this different

Most agencies pick a lane: strategy, design, or development. We worked across all three, plus hardware integration, embedded systems, and organizational change. The project required someone who could talk to mechanical engineers in the morning, refactor Linux boot sequences in the afternoon, and facilitate cross-campus architecture decisions by end of day.

That's not a common combination. But for hardware companies modernizing their software, it's exactly what the work demands.