Layer 1 · Components

Drop finished components straight into your website.

Four SDKs — player, capture, review, report — shipped for Web, iOS and Android, so nobody on your team writes a WebGL viewer or a guided-capture camera flow.

Get API keys Start with the player

4 SDKs · Web, iOS and Android · the API is switchable, a rendered viewer is not

Why this layer

Nobody should have to write a WebGL viewer to buy an inference API.

Every model in Layer 2 needs a complete, well-positioned shot set to read from, and every result needs somewhere to be seen. Both ends are front-end work that has nothing to do with your product's logic — so they ship as SDKs rather than as a specification.

The set

Four SDKs, each droppable on its own.

Take the player without the capture flow, or the capture flow without the report. Nothing here requires adopting the rest of the layer.

Player SDK

Embeddable 360° inspection viewer — drops into your own site against a tokenised report URL

360° exterior: Spin from an indexed frame set, drag and auto-rotate, mobile-first, hotspot-aware
Interior panorama: Cockpit view with navigable points of view
Walkaround video: Playback with timeline markers pinned to detected damages
Photo gallery: Ordered by shot position, before/after slider for raw vs rendered
Hotspot overlay: Consumes viewer/hotspots; tap a damage, get severity and repair cost

Open Player SDK →

Capture SDK

Camera flow that guides a non-expert through a complete shot set, start to finish

Guided shot set: Step-by-step prompts until every required position is covered
Non-expert first: Written for the customer holding the phone, not a trained inspector
Complete or nothing: The flow is what makes the downstream read reliable

Open Capture SDK →

Review SDK

Human-in-the-loop validation and correction of model output

Validate: Confirm or reject each detection before it reaches a decision
Correct: Fix model output in place rather than working around it
Gather: Every correction is more data on the way back in

Open Review SDK →

Report SDK

Composed condition report, customer-branded, web or export

Composed: One document from the whole pipeline, not a pile of endpoint responses
Customer-branded: It reads as your report, not as ours
Web or export: Share a link, or hand over a file

Open Report SDK →

Where it runs

One SDK per target, not one web bundle with wrappers.

The capture flow needs the camera, and the player needs the GPU. Both are platform work, so each target gets a real implementation — including a Capacitor-friendly path, since that is how we run it ourselves.

Web: Framework-agnostic core, plus Vue and React wrappers
iOS: Native SDK
Android: Native SDK
Capacitor: A supported path, because that is how we run it ourselves

What feeds it

These render Layer 2's output. They do not produce it.

Two endpoints do most of the feeding. Both are ordinary calls you can make yourself — the SDKs just consume their responses.

POST /v1/video/onboard
Indexes a walkaround into positioned frames — what the player spins through
POST /v1/viewer/hotspots
Anchors damages and parts to the vehicle — what the hotspot overlay draws

The rest of the stack