Pegasus: AI-Driven Optical Receiving and Product Identification

Make receiving the first point of recovery

Receiving has long been the least instrumented step in the chain. Arrival condition lived in an operator’s eyes, carrier damage was rarely documented per unit, and every downstream decision rested on what was, or was not, captured at the dock.

Pegasus turns intake into a structured decision point. It captures unit identity, condition, and receiving metadata in a single operator pass and writes an image-verified record for every unit. That record becomes the foundation for quality control, carrier claims, refund integrity, and downstream automation.

Identify any product automatically

Receiving teams often handle hundreds of product variants that look nearly identical. The detail that decides the correct part number can be a small mechanical feature, a label variation, a specific QR code, or an identifier hidden beneath an access panel.

Pegasus determines the part number from a product’s visual appearance, serial number, barcode, or label, and when a unit needs a closer look, it guides the operator through a few simple actions to expose the right feature. Operator training collapses to one instruction: present the item to the camera. The operator no longer has to decide which barcode to scan or carry the lookups in their head.

One capture, multiple outcomes

A single computer vision pass produces a structured record that supports several workflows at once, with people kept in the loop for the decisions that matter.

It assembles per-unit carrier-claim evidence by linking the arrival image to order ID, RMA, serial, waybill, and carrier, and ties physical damage to the downstream test result. It raises receiving productivity while reducing scan errors, since the same motion captures and writes every required identifier. And it supports refund and return-policy integrity by detecting missing components at unpackaging and surfacing exceptions immediately.

Built to fit any receiving environment

Full automation is not always practical. Short contracts, fast timelines, and mixed product can make heavy investment hard to justify.

Pegasus is modular. It scales from a single receiving station to a large enterprise deployment, runs as a standalone digital station, and connects to your operating platform through an API. It delivers productivity, consistency, and quality on a manual line, and it grows into a more automated environment when you are ready.

~30 sec

average receiving time per unit

%

reduction in receiving time

1 pass

captures identity, condition, and receiving metadata

Every unit

leaves an image-verified record at intake

AI

product recognition that reads part numbers from appearance

How Pegasus Works

Present

01

The operator presents the item to the camera. That is the workflow. No prior product lookup or specialized training required.

Capture

02

One or more AI-enabled cameras image the unit in its arrival packaging.

Identify and extract

03

Vision models recognize the product and determine the part number from appearance, serial number, barcode, or label, then extract order ID, RMA, serial, waybill or manifest, and carrier.

Assess

04

Pegasus flags damage, missing accessories, and mismatches against the expected manifest, and captures condition evidence tied to the unit.

Record and route

05

Pegasus writes the structured, image-verified record to your operating platform, routes exceptions and claim evidence for review, and resets for the next unit.

Common Deployment Questions

What does a Pegasus station require?
Each station uses one or more AI-enabled cameras, either stationary or portable, a GPU-powered workstation or server to run the AI models and workflow, and a touchscreen interface. AI processing can also run on shared network compute. Optional peripherals include handheld barcode scanners, RFID readers, foot switches, and printers. Hardware cost is roughly $3,000 per station, and existing equipment can often be reused.
What does Pegasus capture at intake?
Pegasus reads order IDs, RMAs, serial numbers, tracking and waybill or manifest data, carrier identification, and product identifiers. It records packaging and accessory condition, captures images of the product and packaging, and, where relevant, flags device lock status. Smart receive controls include RFID and barcode capture, OCR and serial validation, photo evidence capture, and quarantine rule triggers. Everything is stored together as one digital record per unit.
How does product recognition handle near-identical variants?
Pegasus combines vision-language foundation models with proprietary computer vision built for industrial receiving. When the distinguishing detail is subtle, the workflow guides the operator to expose the right feature, so the platform reaches an accurate part number across complex product families.
Which devices and categories does Pegasus support?
Pegasus supports commercial desktops, workstations, notebooks, monitors, accessories, and Chromebooks, along with consumer notebooks and desktops. Additional receiving processes can be configured as needed.
How long does Pegasus take to deploy, and does it need a fully automated line?
Most deployments run about 3 to 4 weeks from purchase order to production, covering workflow configuration and the API connection to your operating platform. Pegasus does not require broader automation. It delivers productivity, consistency, and quality on a manual line and integrates into more automated environments as operations grow.

Built to Work With

Ready to turn receiving into recovery?