Company
Quality control, closed-loop
Seam is built for textile manufacturers and their QC teams — turning AI defect detection into the disposition, traceability and corrective-action workflow a real floor runs on.
Why we built it
Detection was never the hard part
Plenty of tools can draw a box around a defect. What a factory actually needs is the decision that follows — accept or reject this lot, against this customer’s spec, with a record an auditor will accept.
Seam connects detection to disposition, dispositions to corrective action, and every roll to its supplier lot and customer shipment — so quality is a system, not a screenshot.
10
QC workflows
Z1.4
AQL sampling
ASTM
4-point grade
ISO
Audit-ready records
FAQ
Common questions
What defect types can Seam detect?
Seam detects 9 defect classes with YOLOv8: holes, stains, streaks, weft defects, warp defects, needle marks, broken stitches, pinched fabric, and knot/slubs — with tear/cut in active training. Each detection returns a bounding box, class, and confidence. The library grows through the in-app feedback loop, and Enterprise customers can request custom classes.
Does the AI get better over time?
Yes. Every result includes a quick “is this correct?” check. Confirmations take one click; corrections let your team draw the right box and pick the real class. Each correction becomes training data and is queued for the next retrain — accuracy improves on your actual fabric with no manual labeling.
What is lot disposition, AQL and the four-point grade?
Seam grades each lot with the ASTM-style four-point penalty system, runs an ANSI/ASQ Z1.4 acceptance-sampling plan (it tells the inspector the sample size and accept/reject numbers), checks for any banned defect classes in the customer’s spec, and recommends an accept / rework / downgrade / reject disposition. The reviewer signs it off and the decision is recorded with a full audit trail.
Can I trace a roll back to its supplier and forward to a customer?
Yes. Traceability links every roll to its supplier lot and to the customer PO, with the machine, operator and shift it ran on. A single search scopes a recall — “which shipments contain supplier lot X?” — instantly.
How does the AI Analyst work?
It’s powered by Claude Sonnet 4.6, with access to your inspection data filtered to the production line you’re viewing. Ask about defect patterns, root causes or maintenance; conversation history persists across sessions.
How is my data stored?
All images and inspection data are stored securely with Supabase (Postgres + object storage), encrypted in transit and at rest. You can delete individual images or entire batches at any time.
Can I export reports and certificates?
Yes — branded quality certificates (COA) per batch, plus PDF and Excel export and scheduled email digests for client handoffs, supplier audits and internal reviews.
Talk to us
Tell us your fabric type and defect challenges. We’ll show you exactly what Seam catches — on your own images.