Fit drives roughly half of apparel returns, and about 25% of e-commerce items come back — a cost Coresight Research puts at up to two-thirds of product price end-to-end. Rust Coast Manufacturing Technology is betting that inline computer vision can cut that bill by measuring every garment in seconds rather than sampling 1-2% by hand.

Founder Kyle Valade, a former Rivian data engineer, positions the company as a quality platform "that will save companies millions by ensuring correctly sized products are always shipped." Cameras installed across factories photograph garments and run them through visual AI to automatically measure them "comparing them with the spec in real-time," he told Sourcing Journal.

The system feeds photos and data to an app that generates analytics and alerts for the brand's portal. "What we're building towards is having a picture of each piece of clothing all through the manufacturing process," Valade said. "The cameras passively take measurements without getting in the way of the people on the floor."

Why manual measurement fails at scale

A full manual measurement of a pair of pants takes a quality inspector around 20 minutes, according to Valade. That cost forces brands to measure only a subset of an already small inspection sample — typically 1-2% of production or less.

The result is a visibility gap. When consumers return garments for fit, brands cannot tell whether the return reflects consumer preference or a factory defect, because the factory's own report may say every measurement was correct. Valade said this disconnect is what Rust Coast aims to close.

Delayed discovery compounds the problem. With order lead times running to months, a defect found late in production can hit inventory and sales.

How accurate is the system?

Rust Coast measures all visible points of measurement in seconds and lands within one-eighth inch of a manual inspection over 90% of the time. Valade said trained human auditors often diverge from each other by a quarter inch. The company is working to close the gap to 100% accuracy.

Inspector training takes five minutes, and stations work with or without human input. Placement inline in existing workflows is the design principle: competing measurement solutions that add a line step often go unused because floor workers are paid on output.

Real-time measurement also changes factory decision-making. If a batch of pants measures closer to a different size than intended, the factory can relabel, repair or reject the batch before adding materials and shipping cost. Defects grow more expensive at every downstream stage, Valade noted — catching a spec error immediately after sewing beats failing final inspection after full make-up.

What does this change for sourcing teams?

The platform does not replace human inspectors; it augments them, with the potential to inspect 100% of garments inline. For quality teams stretched across supply bases in multiple countries, the value is data: images and measurement statistics for every tracked garment, replaceable with high-level factory summaries.

Rust Coast is building toward digital product passport capability and supplier scorecard integration. Connecting returns data to production records would let a brand scan the sewn-in tag of a returned garment, pull up images and measurements from its production batch, and determine the blast radius of a consumer-flagged issue — one item, one batch, or a full SKU across the supply chain.

Valade gave a concrete use case: "We're getting some complaints around fit for this color of denim. Looks like it's shrinking in the wash more than expected — let's tell our suppliers so they can account for that."

He frames the savings in hard numbers: reducing fit-driven returns by even 10% could save a larger brand millions of dollars.

For sourcing and quality executives, the decision Rust Coast forces is whether to move beyond statistical sampling toward full inline measurement data as the basis of supplier oversight — and whether the return-cost math justifies adding vision stations across a multi-country factory base.