Rust Coast Manufacturing Technology claims its camera-based visual AI can measure garments in seconds and accurate to within one-eighth of an inch over 90 percent of the time, against a 20-minute manual measurement — a proposition aimed squarely at the fit problem that drives half of apparel e-commerce returns.

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

Photos and measurement data flow to an app for deeper processing that generates analytics and alerts into a brand 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."

What problem is Rust Coast attacking?

The company started not with measurement but with digitizing pencil-and-paper factory processes. Valade found that even simple products — he cites beanies — require multiple steps where a single error ruins the garment, from knitting and sewing through labeling and calibration.

At Rivian, Valade helped organize petabytes of vehicle data used for factory inspection. Automotive and other industries run automated inspections; apparel largely does not. Brands sourcing from many factories across many countries rely on tech packs and manual checks that take time and miss variables.

The commercial logic rests on returns. About a quarter of e-commerce items come back, most for fit reasons, and Coresight Research data cited by Valade puts the end-to-end cost of handling a return at up to two-thirds of product cost. Reducing fit returns by even 10 percent, he argues, "could easily save millions of dollars for a larger brand."

How does the inspection economics change?

Sampling reflects an assumption: inspecting everything is too expensive. Factories typically check 1 to 2 percent of items or less, accepting the risk that the sample misses problems. Rust Coast does not claim to replace inspectors. It augments them, and placed inline it can inspect 100 percent of garments — "maybe not as thoroughly as a person—but far better than no inspection at all," Valade said.

Early defect detection carries direct cost weight. Defects grow more expensive to fix further along the production line; catching pants that run tight immediately after sewing avoids pouring more materials and labor into goods that fail inspection later. Real-time measurement also opens operational options: a batch closer to a different size's spec could be relabeled, repaired, or rejected before shipping costs mount.

Deployment friction is a stated design constraint. Competing measurement tools often add a line step, a cost factories resist where floor workers are paid on output. Rust Coast stations work with or without inspector input, position across multiple stations, and inspector training has taken five minutes, according to the company.

What visibility do brands gain?

Beyond returns, reviews and high-level factory quality summaries, brands currently have little product data, Valade argues. The platform aims to produce an end-to-end record per garment across its lifecycle, feeding supplier scorecards rather than periodic summaries.

Returns integration enables root-cause tracing. Brands can key a returned garment's sewn-in purchase order number into the app and pull production images and measurement data for that batch, defining the "blast radius" of a consumer-flagged issue — one defective item, one batch, or a full SKU across the supply chain.

Valade gives a concrete example: complaints around fit on a specific denim color could be traced to unexpected shrinkage, prompting corrective instruction to suppliers. Quality teams, he says, can shift from measurement and visible defects toward preventive process work — a meaningful reallocation for teams that "can't be in twelve countries at once."

What remains to be proven?

The company's own framing separates ambition from demonstrated result. Goals are stated as goals: measuring 100 percent of garments, pushing accuracy from over 90 percent at one-eighth-inch tolerance toward 100 percent, and a full per-garment lifecycle record. Measured claims so far are the accuracy benchmark — against a one-quarter-inch discrepancy between trained human auditors — and the five-minute training time.

For sourcing and quality executives, the immediate decision points are whether inline 100 percent measurement beats statistical sampling on cost, how the stations fit existing line workflows, and who pays for installation across a multi-factory vendor base. Valade positions the data layer as the longer-term play, feeding supplier scorecards and future AI initiatives, with the pitch that finding defects earlier in the process saves factories money and cuts inspection time significantly.