CS—08 · Conglomerate · Data & AI
Qimma — Group Data & AI Platform
Fourteen business units, fourteen spreadsheet pipelines — consolidated into one governed lakehouse with Arabic-capable document intelligence.
A diversified Saudi holding group · 16-month program
The challenge
Fourteen business units reported through fourteen spreadsheet pipelines. Group finance closed the month three weeks late, and no one could answer cross-unit questions about customers or suppliers.
Contracts and correspondence — the group's institutional memory — sat unsearchable in Arabic-language PDFs.
What we built
We built a governed lakehouse with unit-level data contracts and quality gates, migrated reporting to certified semantic models, and put finance, operations, and the board on the same numbers.
Arabic-capable NLP now classifies contracts and customer feedback across the group's markets, and a retrieval-grounded assistant answers questions against the governed data — with access controls intact.
Next case study
Meras — Core Banking Modernization
Let's build what's next. لنبنِ المستقبل معاً
The Kingdom's IT market is racing from $20 billion to $45 billion by 2030 — and your digital roadmap decides your share of it. Tell us where your organization is headed; we'll bring an architecture, a plan, and a team that has done it before, within one week of your brief.