I built a sourcing copilot because "premium feel" isn't a technical spec

I've been working in sourcing and procurement for a while, and there's one part of the job that consistently eats my time: the gap between what a customer asks for and what a factory needs to hear.

Last month, a stakeholder sent me a product brief that said: "Need a sturdy travel mug, premium feel, holds coffee hot for a while, our logo on the side, ready for our holiday launch."

That's it. No dimensions. No material. No tolerance. No packaging requirements.

I spent 3 hours turning that into a 12-section technical spec sheet before I could even send an RFQ to suppliers. Then I had to manually compare quotes across different formats, and rebuild my landed cost spreadsheet because Section 301 tariffs changed again.

I got fed up and built Spec AI — a lightweight, web-based sourcing copilot with four tools:

1. Spec Translator — Turns vague customer briefs into technical spec sheets with tolerances and flagged ambiguities.

2. RFQ Engine — Generates comprehensive RFQs from a one-line plain English description. Materials, tolerances, packaging, supplier questions — all included.

3. Supplier Matrix — Paste messy supplier notes (quotes, lead times, MOQs, certs) in any format. It builds a scored comparison and picks a winner.

4. Landed Cost Co-Pilot — Deterministic calculator. You enter the HTS code and tariff stack; it handles Section 301, Section 232, MPF (with min/max caps), HMF, and every shipment fee to 4 decimal places.

Current status: Live, free, and I'm the only user. I built this as a side project while working my day job.

What I need: I'm trying to figure out if this is a real pain point for other procurement/sourcing folks, or if I just have a weird tolerance for manual work. If you work in sourcing, manufacturing, or logistics and want to kick the tires, I'd love your honest feedback.

Drop a comment and I'll send over the link. Or just tell me why this won't work — I'll take that too.