AeroQuote Compass turns a stack of supplier quotations into an auditable sourcing decision. Buyers upload quotes in any format, AI extracts the commercial terms, the buyer reviews and confirms every field, and deterministic code calculates total annual landed cost, supplier scores, risks and a justified recommendation.
The problem
Quotes for the same RFQ rarely compare like for like: different currencies, Incoterms, price breaks, freight and insurance, tooling, NRE, part revisions and validity dates. Choosing on unit price can award the wrong supplier and hide compliance gaps.
How it works?
1. RFQ setup: part number, revision, annual volume, required certification, FX rates, target cost and incumbent supplier.
2. Upload quotes: up to 20 files per RFQ (PDF, Excel, CSV, TXT) or pasted text.
3. Review data: AI extraction with a confidence score per field. Nothing is calculated before the buyer confirms.
4. Compare and decide: landed cost comparison, configurable supplier scoring, risk and anomaly flags, recommendation and export to PDF, Excel or CSV.
Design principle: AI extracts, buyers validate, code calculates. The LLM never produces a number that drives the decision, and every score shows its formula.
What it catches?
In the sample RFQ, the incumbent had the lowest unit price and the highest annual landed cost once freight, insurance and NRE were included. The two cheapest landed offers were not awardable: one quoted the wrong part revision, the other had expired. The recommended supplier saves 3.2% per year versus the incumbent.
Who it is for?
Procurement and strategic sourcing teams, buyers comparing international quotes, and anyone who has ever normalized supplier quotes by hand in a spreadsheet.
All data is synthetic and supplier names are fictitious. Built by an industrial engineer with almost five years of experience in aerospace procurement.
Source code: github.com/stephanyalves-eng/aeroquote-compass
Built with