Mecaniq

If you've ever stood next to a piece of factory equipment and known — known, before anything was on a screen — that the bearing was about to go, you already understand what Mecaniq is for.

The maintenance lead at any well-run plant in America has this kind of knowledge. They've spent ten or fifteen years walking the same line, running their hand along the same housings, listening to the same motors come up to speed and shut down. They've heard the noise the centrifugal pump makes the week before the seal fails. They've felt the warmth on the gearbox housing two weeks before the tooth chips. Some of what they know they could explain. Most of it they couldn't, because most of it lives in pattern recognition that has accumulated below the level of language.

That knowledge is the most valuable asset on the plant floor, and most of the software being sold to manufacturing right now treats it as an obstacle. The pitch from the predictive-maintenance vendor goes: install our sensors, install our dashboard, and stop relying on tribal knowledge. The plant takes the meeting, signs the paperwork, deploys the tooling, and a year later the dashboard is the tab nobody has open. The maintenance lead is still walking the line. The senior machinist is still running their hand along the spindle housing. The software is still on the budget but no longer in the workflow.

Mecaniq starts somewhere different. The premise is that the maintenance lead's pattern recognition is the asset, and the platform's job is to capture, share, and extend it — not to replace it. The mécanique part of the name is the engineering substrate: pumps, motors, bearings, drives, controllers, valves, sensors, the whole class of mechanical and mechatronic systems that produce the work in any plant. The IQ part is the intelligence overlay: the vibration trace, the load-current signature, the temperature drift, the cycling pattern — the data that's already being produced by the equipment and that, in most plants, is going almost entirely unused.

The platform reads what's there. Modern PLCs, legacy controllers from the 1990s, retrofit sensor packs on equipment that doesn't have native telemetry, even hand-logged values the old-timer writes on a clipboard. It normalizes across the equipment, classifies anomalies in the language the maintenance team already uses, and produces three things the existing tools don't reliably produce: failure forecasts that arrive before the failure, condition-based maintenance schedules that survive contact with a real shift schedule, and operator-facing playbooks written in the voice of a senior practitioner.

The voice on the platform matters because the buyer matters. Mecaniq isn't sold to a chief data officer or to an enterprise IT committee. It's sold to the plant engineer, the reliability lead, the maintenance director — people whose pattern recognition has already done the hard work of figuring out which assets fail in which ways, and who are looking for a tool that respects the practical knowledge they've built up over careers. The interface reflects that respect. Asset cards are named in plant vocabulary, not in serial numbers. Alerts come with playbooks, not just severities. The 'acknowledged' state on every notification is built around the assumption that the human is the source of truth, not the algorithm.

Mecaniq is built for the mid-market plant, not the supermajor. The large enterprise predictive-maintenance vendors are credible — Augury, Uptake, GE Digital, Siemens MindSphere — and they serve customers who can afford to staff a full reliability team and absorb a six- or seven-figure platform contract. That's not who Mecaniq is for. The mid-market manufacturer — fifty to five hundred employees, one reliability engineer, three operators per shift, a maintenance budget that has to make the math work — is the audience the platform was designed around. The product, the pricing, the implementation playbook, the support model — all of them are sized to that operator, and none of them are afterthoughts to a different buyer.

The category framing matters too. Most predictive-maintenance vendors specialize: vibration-only, motor-current-only, lubrication-only, oil-analysis-only. Real machinery is mechatronic, which is to say mechanical, electrical, and computational layers running together — and a fault in one layer almost always shows up in another layer first. A failing bearing telegraphs in vibration first, in motor current second, in temperature third. The platform that monitors only one of those signals will catch the failure on the third reading; the platform that watches all three will catch it on the first. Mecaniq treats the mechatronic join as the unit of analysis, which is a small distinction in vendor pitches and a large one in actual signal-to-noise.

The brand is built around all of it: the audience, the framing, the voice, the visual identity. The mark is plain and quiet — the wordmark in Nunito, the body in Lora, the accent a single slate-teal on near-white. The visual posture is closer to a lab notebook than to a SaaS dashboard, because the buyer is closer to a craftsperson than to an enterprise procurement officer. None of this is decoration. It's all in service of the same idea: that the maintenance lead's pattern recognition deserves a platform that respects it, and the rest of the manufacturing economy is overdue for one.

Mecaniq. What your machines already know.