The demo below runs on sample data.
RFQs parsed from PDF and Excel into editable quotes
RFQFlow. An RFQ platform that reads PDF and Excel tenders, extracts line items and specs, and moves each quote through comparison and approval in one place.
- Client
- An industrial supplier quoting engineering tenders
- Location
- UAE / GCC
- Year
- 2026
- Services
- AI Agents & Workflow Automation, AI Development, Data & Analytics
- Industry
- Quality, MES and inspection software for manufacturers
- Stack
- Next.js, React, TypeScript, Tailwind CSS, OpenAI
- Status
- Interactive demo
Overview
ShichifukuTekx built an AI-powered RFQ platform for an industrial supplier. It parses PDF and Excel RFQ documents into line items, quantities and specifications, tracks every quote on one dashboard, compares supplier offers on price, lead time and quality history, and routes high-value quotes through configurable approval chains before they go out.
Where they started
- 01
RFQs arrived as PDF packs and Excel sheets in different layouts. Estimators retyped every line item before they could start pricing, and typing errors carried straight into quotes.
- 02
Engineering tender packages ran to several hundred pages of drawings, specifications and commercial terms. Lines were easy to miss.
- 03
Active RFQs, pending quotes and awarded contracts were tracked in separate spreadsheets. Sales leadership had no single view of the pipeline.
- 04
Supplier prices, lead times and past performance sat in inboxes. Comparing offers meant rebuilding the same table for every tender.
- 05
High-value quotes needed management sign-off, but approvals happened over email with no record of who approved what.
What we built
We ran a discovery sprint on real tender packs to understand the layouts, units and specification formats the team receives most often. Those samples shaped the extraction rules and the review screens.
- AI parsinggenerative AI extracts line items, quantities and specifications from PDF and Excel RFQs
- Pipeline dashboardactive RFQs, pending quotes and awarded contracts in one view
- Quote comparisonside-by-side supplier offers, with best-value options highlighted on price, lead time and quality history
- Supplier databaseperformance ratings, approvals and historical pricing per supplier
- Approval workflowsconfigurable chains route high-value quotes to management
- Cost analysisspending patterns and supplier cost trends over time
- Structured BOM outputextracted items grouped into BOM lines with quantities and descriptions, mapped to the company's own cost codes
Parsed lines land in an editable quote. Each line stays linked to the source document, and low-confidence extractions are flagged for review. The estimator checks the extraction, adjusts where needed and stays in control of what goes to the customer.
The platform is built on Next.js, React and TypeScript with OpenAI models for document understanding. Analytics dashboards show quote volume, win and loss trends and the share of RFQs processed through AI parsing.
The platform runs today as RFQFlow at rfqflow.shichifukutekx.ai, used by procurement and tender teams to turn tender packages into structured BOMs and vendor comparisons on a common basis.
What changed
- PDF + Excel, rFQ formats parsed. Line items, quantities and specs extracted into an editable quote
- Side-by-side, quote comparison. Price, lead time and quality history in one table
- Configurable, approval chains. High-value quotes route to management with a recorded sign-off
- One pipeline, from RFQ to award. Active, pending and awarded work tracked on one dashboard