W-Enterprises/Products
Product studio
Every tool on this page shipped inside a real engagement before it became a product: it screened real marks, parsed real invoices, generated real HR documents, answered real questions with citations. What you get is not a roadmap. It is running software with a pilot path.
Each product exists because a paying engagement forced it into existence, and each carries the same doctrine: bilingual where it matters, nothing sensitive persisted by default, every claim measurable, and the option to run on your own infrastructure.
Trademark clearance screening across four independent axes - appearance, wordmark sight and sound, palette, device geometry - where each axis carries its own decision ceiling, so one weak signal cannot average away a real conflict. Generated alternative marks are re-screened against every registration scored, and every published figure is pinned to a recorded run with config and source hashes.
The full mechanism and demo access →
OCR plus LLM normalisation for invoices and forms in Arabic and English: about 93.5 percent field accuracy on the shipped MVP scope, a draft-approve-push review workflow instead of blind trust, five ingestion channels including WhatsApp, and nothing persisted to disk by default.
How it works → The case study →
Branded bilingual document generation: Arabic and English templates with correct right-to-left rendering, server-rendered PDFs, theming per tenant, and no field values persisted. The engine behind a live multi-tenant HR document platform with 34 templates and Geidea payments in front of it.
The case study →
Ask questions across your own files and get answers with citations, not vibes. Sources are scanned incrementally into Postgres with pgvector; every chunk carries a sensitivity classification; private material requires an unlocked private session; cloud calls are off by default. It runs daily inside the practice's own operations cockpit.
Request a pilot →
Products here are bought the way the consulting is bought: in writing, with honest scope. The difference is that the system already exists, so a pilot starts from running software instead of a blank repository.
A short, fixed-scope run against your real data: your marks, your invoices, your templates, or your document corpus. Success criteria are agreed in writing before it starts, the price is fixed, and you see measured results on your own material, not brochure numbers.
Yes, and for regulated or Gulf deployments that is the default posture: your cloud account, in-region hosting where required, and a no-persistence configuration for sensitive fields. You are not asked to trust a third-party multi-tenant box with regulated data.
The default across all four products is that documents stream through and are not written to disk, audit trails record metadata rather than field values, and anything private stays behind explicit gating. That posture is built in, because the products were built for PDPL, POPIA and GDPR environments first.
Clearance Studio has a live demo; access codes are granted on request, usually the same day. The extraction and document engines are demonstrated live on a call against sample documents. The knowledge layer is piloted directly on a scoped corpus you choose.
Occasional, substantive updates when a product ships, opens for pilots, or publishes new measured results. No drip sequences.
Say which product fits and what you would run it against. You will get a written read on fit, pilot scope and price within one working day.