Mordéa Home × Seamora AI
A live view of the Mordéa Home operations system as it's being built: overall progress, every milestone shipped, and the current state of each module. This page updates automatically as development moves forward.
46 engineering tasks: 22 done, 3 in progress, 21 queued.
Sign-in & roles, dashboard, database groundwork · 17/25 done
5/15 done · 1 in development
0/6 done
One entry per shipped milestone, newest first.
The factory-facing portal shipped its first version: a factory signs in and sees just its own production orders and shipment status — fully de-identified, with buyer details and retail prices structurally removed from that view.
The system now keeps a profile for every factory — contacts, account details, partnership status — and portal accounts can be created and invited straight from a profile page.
The orders page took its back-office shape: dense table, filters, drawer detail views — every purchase order (PO) on one screen — plus the PO→FO conversion wizard: pick lines, allocate quantities to factories, and track consolidated or split shipments through a proper status flow. Manual entry stays the baseline, so the system works standalone; Excel parsing and AI capture plug in later purely as accelerators.
The production database is now running in the company's own dedicated AWS account, hosted in Sydney to meet the requirement that data stays in Australia. Sign-in user data has been migrated across. Every operation leaves an audit trail, and at handover the whole account can be transferred as-is.
All 17 under-construction pages were replaced with interactive demo pages filled with sample data — the whole system can now be clicked through end to end, so flows get validated before real data lands.
The system now uses proper email-and-password sign-in, with four roles — owner, manager, designer, supplier — controlling what each person can see, down to individual pages and fields. The temporary password gate is gone and demo accounts are ready.
First module delivered: the business dashboard runs on de-identified real business data — sales, orders and channel KPIs, every number labelled with its source and as-of date — plus drill-down order lists and detail pages with polished loading, hover and keyboard interactions.
Project kickoff deliverable: the app shell went live with the brand design system applied and navigation in place for every planned module, and a first full dashboard concept page was reviewed for layout and information architecture — before any real data went in.