Operations Command Center
National customs authority · West Africa
A live executive command center for a national port, joining five production databases into a single operational picture.
- React 19
- TypeScript
- Vite
- Tailwind
- +4
Customs clearance, port operations, national registries — and the machine learning and verifiable ledgers running inside them. We build for institutions where being wrong is expensive, and being unable to prove otherwise is worse.
Our clearance platform runs at three national customs administrations. Each deployment is configured rather than rebuilt, so an authority adopting it inherits everything the previous two paid to discover.
That makes a fourth deployment a programme with a known shape and a known cost — not a bespoke quote assembled from guesses.
Discuss a deployment →Capture, OCR intake, axle compliance, custody
Declarations, duty waivers, review and release
Appointments, exit control, terminal registries
ASYCUDA, ICUMS and UNIPASS document mapping
Payments, targets, variance and reporting
Executive dashboard across every module
The mechanism repeats more than people expect. A customs clearance chain, a student record and a patient referral are the same problem wearing different clothes: several parties, one record, and a decision that has to be defensible afterwards.
Systems where the auditor matters as much as the user, and where the record has to survive being challenged years later.
Institutions running admissions, enrolment, assessment and fees across several departments that each believe they own the student record.
Clinical and administrative systems where access control, auditability and uptime are not features but preconditions.
Regulated environments where every balance change needs an attributable trail, and reconciliation is the product rather than a report.
Moving goods through checkpoints where several independent parties must agree on one custody history.
Customer-facing brand and booking platforms, built to stay fast where bandwidth is expensive.
Most failed public-sector software fails before a line is written — scoped from a document instead of from the work. We start on the floor, with the people who will actually use it.
The full engagement model →We sit with the people doing the work and map the process as it actually runs — not as the manual describes it.
Architecture, scope and a fixed price — produced from what we observed, never from the document alone.
We build the smallest thing that proves the mechanism, put it in front of real operators, and measure it.
Production means monitored, documented, access-controlled and backed up — verified, not asserted.
Source, runbook and training. We stay because you want us to, not because you cannot operate without us.
Start here
A 30-minute technical briefing. We will tell you what we would build, roughly what it costs, and honestly whether we are the right people for it.