Home / About

About

Built for the line, not the lab

tyle.tech is a manufacturing-AI company for ceramic and porcelain tile producers. We put computer vision and process models directly on the production line — where a defect is cheap to catch and expensive to miss.

Tile is one of the oldest manufactured products in the world, and one of the least forgiving. A batch that’s a little too wet, a die that’s a little worn, a kiln that runs a few degrees hot — none of it announces itself. It shows up hours later as a cracked, warped, or off-shade tile that already cost you clay, glaze, and gas to make.

We started tyle.tech because the tools meant to catch these problems were built for the lab, not the line. Sampling a handful of tiles an hour, or reviewing defects after firing, tells you what went wrong long after you could have fixed it. We think inspection belongs where the tiles are moving — at belt speed, on every piece, with a model watching each stage rather than a person watching a monitor.

Our team comes from computer vision, industrial automation, and the ceramics industry itself. We deploy onto existing belts and PLCs, learn your line’s own definition of “good,” and give you back yield, energy, and a lot fewer surprises at the sorting table.

By the numbers

2021
Founded
Built by vision, automation, and ceramics people.
40+
Lines deployed
Across ceramic and porcelain producers.
1.2B m²
Tile inspected
Every piece, not a sampled tenth.
100%
On-line
Inspection at belt speed, not in a lab.

How we work

Three things we hold to

On the line

If a model can’t run at belt speed on your existing hardware, it doesn’t help you. We deploy where the tiles are, not in a report you read tomorrow.

Your definition of good

Every plant grades differently. We calibrate to your specs and lots rather than imposing a generic standard from somewhere else.

No rip-and-replace

We add to the line you already run — belts, kilns, PLCs, and all. No rebuild, no multi-week stoppage to get value.

Line audit

Bring us one week of your reject data.

We’ll map where your yield is leaking and model the lift before you commit to anything. Most audits pay for the first quarter of deployment.