Real Projects.
Real Numbers, Where We Can Share Them.

The challenge

A major German health insurance auditing organisation was processing around 15 million prescriptions a month, with a team of roughly 200 staff responsible for catching fraudulent or altered claims. Manually, they could only spot-check around 200,000 of those — under 2% — leaving the rest unverified.

The approach

Dr. Seify designed and piloted an image-processing, AI-powered verification platform that reads each scanned prescription and checks it against the authenticity markers a genuine document should carry — stamp, signature and encryption/security features — flagging anything inconsistent as a likely fraud.

The outcome

Instead of a ~2% manual sample, the pilot platform was able to check effectively all 15 million monthly prescriptions in around five days — at validated accuracy of approximately 95%, far faster and broader than the manual process it was tested against.

The challenge

Unilever's formulation team used a robotic arm to test different toothpaste formulations, pressing a calibrated tool into paste samples and photographing the resulting indentation to judge consistency. The problem: there was no reliable way to match each photo back to the exact tool-and-formulation combination that produced it, which broke the ability to compare results at scale. An external manufacturing consultancy quoted approximately £150,000 to fix it.

The approach

Dr. Seify diagnosed the root cause of the data-matching failure and engineered a lightweight fix that reliably linked each image back to its source formulation and tool — without the large-scale rebuild the original quote assumed was necessary.

The outcome

"What sets VisionXY7 apart is the combination of technical depth and genuine ethical rigour. They never pushed a solution that wasn't the right fit." — Matt Tynan, Unilever

The challenge

A hospital haematology laboratory processes around 2,000 blood specimens a day through its automated blood analysers, under constant pressure to meet daily clinical demand without excess energy and equipment cost.

The approach

Dr. Seify designed and deployed an IoT monitoring platform across the laboratory, running for several months and collecting well over a million operational data points. This work fed directly into the doctoral Deep Business Analytics (DBA) framework, later validated at the Royal Liverpool Hospital NHS Haematology Laboratory against more than 50 million clinical records.

The outcome

AI-driven analysis of the monitoring data identified the operational arrangement that minimised energy consumption while maintaining the laboratory's ability to meet full daily testing demand.

The challenge

A $200M World Bank-funded, multi-ministry development programme needed a single point of monitoring, evaluation and management-information coordination across several government ministries and international consulting firms — each with its own reporting standards, timelines and priorities.

The approach

As Strategic M&E and MIS Officer, Dr. Seify built and ran the programme's coordination and alignment function — bringing 3 government ministries and 5 international consulting firms onto a common reporting and governance structure so the programme could be tracked and steered as one.

The outcome

This is the same governance discipline — coordinating multiple stakeholders around one accountable structure — that now underpins VisionXY7's Enterprise AI Governance and Fractional Chief AI Officer engagements.