TexPro — Market Intelligence Rebuild + Moda AI
Fibre2Fashion's paid market-intelligence dashboard for the textile industry, rebuilt from the ground up in Angular 21 — trade, pricing, sustainability and supplier data — with Moda AI, a full-page assistant that streams answers from the dashboard's own data modules.
- Year
- 2026
- Type
- Day job
- Role
- One of three core developers · Angular 21
- Status
- Live
What needed solving
TexPro sells trade flows, raw-material prices, country costs, sustainability and supplier data to textile businesses. Its legacy front end carried two generations of code side by side — the chatbot service alone kept both, call by call — so it is being rewritten from scratch and ported module by module against the live APIs. My TexPro work started a year earlier at the other end of the same data: client-specific trade reports, built per subscriber.
The solution
Three of us are rebuilding the dashboard as a zoneless, signals-based Angular 21 app from Figma designs. Most of my share is Sustainability (Energy Statistics, GHG Emission, Brand Targets) and Supplier Network (Supplier List, analysis by brand and by geography), with a large part of Global Trade Data, Raw Material Prices and Tariff Duties, on a shared Highcharts and virtualised-table layer. I led Moda AI's front end: a full-page assistant streaming NDJSON from the Copilot endpoints, with a conversation history rail and suggested questions, opened only where the reader's plan and modules allow it. Before the rebuild, I wrote 16 of TexPro's client-specific reports in C# and EPPlus — Excel workbooks built from the trade data and emailed to each client with a PDF copy.
What changed
- Moda AI built to the approved Figma design: streaming answers, day-grouped history with rename and delete, and suggested questions over the modules it answers from
- The AI Insight panel's Ask Anything chip, disabled since the panel was built, now opens a per-document chat in a side drawer
- Sustainability and Supplier Network — six screens between them — are where most of my rebuild work went, with Global Trade Data, Raw Material Prices and Tariff Duties alongside
- Earlier: 16 client-specific trade reports in C# and EPPlus, delivered to subscribers as Excel and PDF
Technical highlights
Need something like this?
I take on a small number of projects each quarter. Let's talk if your idea fits.
News Desk — Human-in-the-Loop Story-Discovery Agent
An automated newsroom agent that reproduces an editorial team's morning routine — harvests ~65 primary and ~75 secondary curated source pages across 17 countries for textile/apparel news, scores and de-dupes the best stories with an LLM, confirms each against its primary source, then drafts publish-ready articles. Two-phase with a human in the loop, streamed live over SSE.
AI Article Gen — Document-to-Publish Pipeline
Turns a raw Word document into a publish-ready article package — headline, key takeaways, table of contents, body, SEO metadata, FAQs — plus the DOCX, PDF and HTM artefacts the public site serves.