Case studiesRetail operations

PLEXOR: connecting retail video and point-of-sale events

PLEXOR connected the information in retail camera feeds with the transactions at the register. Zain Raza led the backend and AI platform build for Ego Eimi, turning those inputs into events a manager could review.

Business
PLEXOR
Sector
Retail operations
Scope
Video and point-of-sale event analysis
Contributor
Zain Raza
Editorial illustration: a retail camera and receipt roller connected by an orange paper strip
AI-generated editorial illustration of the project’s domain.

A recording and a receipt showed different parts of an event.

Retail video and point-of-sale records often sit in separate systems. Investigating a transaction means finding the matching moment in a recording, while other operational events may have no corresponding receipt.

The product needed to bring those signals together and describe a potential event clearly enough for a manager to examine it. The existing IP camera infrastructure was an input to that design.

Join video context to transaction data.

The backend combined live video with transaction data. Vision-language analysis interpreted events and produced plain-language descriptions connected to their context.

The application supported review of potential loss, service and safety events. Linking the footage and transaction information let a manager examine the underlying evidence rather than relying only on an alert label.

What the workflow includes

  • Ingest existing IP camera feeds and POS records.
  • Associate a transaction with its video context.
  • Use vision-language analysis to describe potential events.
  • Present event context for a manager’s review.

Project details

The PLEXOR system architecture diagram showing existing IP cameras and POS feeding a Vision-Language AI that produces real-time alerts

Project materials showing the workflows and interfaces described in this case.

Product walkthrough

Video and point-of-sale event analysis. The delivery section explains the workflow demonstrated here.

Managers could inspect an event with its supporting context.

The delivery connected retail observation and transaction data in one analysis platform. The product’s engineering contribution was the route from separate inputs to an inspectable event, with a person responsible for interpreting what it means.

What this means for the technology owner.

A video-analysis system needs explicit access, retention and review rules. An AI alert should not become an automatic accusation or employment decision. The operating scope should cover how footage is handled, how incorrect alerts are reviewed and who can act on the information.

How we operate AI workflows →

Vision-Language AIComputer visionIP camerasPOS integrationReal-time inferenceCloud

Project contribution: Zain Raza. The scope described above identifies the work behind this case.

PLEXOR — project link

Editorial review: Ego Eimi · Updated September 6, 2026 · About the team

Your technology

Start with the business

Your business depends on it.

Tell us which system matters, what has changed and who controls it. We will define a reviewable next step. Read Foundation and the commitments and evidence.