Product
Every line item traces back to a photo
PaxExact records damage as evidence-linked observations, not a freeform scope narrative, so a reviewer can follow any line item back to the exact photo and measurement that support it.
What's real today
- Photo capture is live: photos attach to specific rooms and damage records from a web browser today; there's no dedicated mobile app yet.
- Evidence-linked scope records are live: line items, measurements, and photos are stored as separate, linked records — not one flat description.
- AI-assisted damage observations are in beta: a photo can be analyzed into a proposed observation tied to that exact photo, and it always requires human review before it counts toward anything.
- Suggested line items are in beta: an accepted observation can be turned into suggested line items with quantities, each of which a professional must approve before it counts. Specific repair operations are suggested only for water damage to walls, ceilings, floors and baseboards, and smoke, soot or mold damage to walls, ceilings and floors.
Evidence is the record, not an attachment
Most estimating tools treat photos as attachments — proof filed alongside an estimate that someone else typed. PaxExact treats them as the record itself. A photo is captured against a specific room and a specific damage entry, and everything built afterward — the observation, the measurement, the line item — points back to it.
That capture happens in a web browser today. There's no dedicated mobile app; it's on the roadmap, not in early access. A LiDAR room scan made in another app can be uploaded as evidence on a room (in beta — see LiDAR & spatial capture). If your workflow depends on capturing evidence from a phone in the field, plan around a browser today and treat a native mobile app as a future capability, not a current one.
What the AI actually does today
AI-assisted damage observations are in beta: PaxExact can analyze a photo and propose a damage observation tied to that specific image. It's real and in use, but it's deliberately narrow. If there's no photo, there's no proposal — the system fails safely instead of guessing at a damage type from context or a description.
Every proposal requires a professional to review it before it counts toward anything. Accept it, edit it, or discard it — none of those paths happen automatically, and none of them touch the estimate until a person signs off.
How a proposal fails safely
From observation to line item — suggested, then approved
The data model for evidence-linked scope is complete and in production: a line item, a measurement, and a photo are separate, linked rows you can trace end to end. The handoff from an accepted observation to a line item is in beta: PaxExact can turn an accepted observation — AI-proposed from a photo or recorded by hand — into suggested line items, with the quantity taken from the observation's own extent and the link to its evidence kept.
The scope is limited: specific repair operations are suggested for water damage to walls, ceilings, floors and baseboards, and smoke, soot or mold damage to walls, ceilings and floors (water damage to a floor does not include replacing the floor covering); any other observation becomes a line with no code or price unless exactly one code fits. A suggested line counts toward nothing until a professional approves it, and the estimator can still add any line by hand.
Why this matters when you defend an estimate
A carrier reviewer, or an adjuster defending an estimate months later, doesn't want a narrative — they want to know exactly what a number is based on. Because photos, observations, measurements, and line items are linked records rather than a single document assembled after the fact, that trace exists by construction, not by someone remembering to keep good notes.
FAQ
Frequently asked
Does AI decide what's damaged?
No. Photo analysis proposes an observation tied to a specific photo; a professional must review and accept it before it becomes part of the record. If there's no photo, there's no proposal — it fails safely rather than guessing.
Are line items generated automatically from photos?
Not automatically, and never without approval. In beta, an accepted observation can be turned into suggested line items — specific repair operations for water damage to walls, ceilings, floors and baseboards, and smoke, soot or mold damage to walls, ceilings and floors — and each one counts only after a professional approves it.
Is there a mobile app for capture?
Not today. Photo and measurement capture happens in a web browser. A dedicated mobile app with built-in LiDAR room scanning is planned but not built; a scan made in another app can be uploaded from the web, in beta.
Related