NADIR / Company / 01 of 04
A monitoring layer for the thing everyone assumes was done correctly.
Roughly two thirds of collision repair orders touch ADAS components, and a substantial share of the required recalibrations are skipped or left incomplete. Nobody finds out until something happens. NADIR exists to shorten that interval.
The thesis
The gap is the window, not the calibration.
Calibration bays are good at what they do. The problem is not that the measurement is wrong; it is that the measurement happens once, in a controlled room, and then the vehicle goes back onto a cambered road with a load in it and thermal cycling and vibration, and nobody measures it again until the next scheduled inspection.
On a monitored fleet a first flag lands at 41 hours after the repair order closes, a bay is booked at 58, and the event closes signed at 72. Unmonitored, the same vehicle is found at the next scheduled inspection — 690 hours later. That interval, not the calibration itself, is the product.
Everything else follows from being disciplined about that one claim. We are not building a better bay, a better sensor, or an autonomy stack. We are building the layer that notices, and the evidence that lets somebody act on it.
Constraints
The things we decided not to do.
NADIR has no ECU write path. Not a disabled one, not a permissioned one — there is no actuation code in the product. That single constraint is what allows a monitoring layer to be deployed on a working fleet without a functional-safety case, and it is the reason the honest word for the output is advisory.
We do not operate calibration bays and do not intend to. We do not sell a hardware box. We do not retain raw telemetry beyond thirty days. And we do not price capabilities that are not built, which is why the roadmap is published with the unbuilt half drawn as outline.
Stage
Where we actually are.
NADIR is early. Two founders, a working detection stack, an evidence pipeline that verifies offline, and pilots scoped around a single agreed metric. We are onboarding fleets and repair networks a few at a time because the constraint right now is attention rather than capacity, and a pilot that nobody is watching closely is a pilot that teaches you nothing.
If you run a fleet, a collision network or an insurance book and the post-repair window is a real problem for you rather than a theoretical one, the useful next step is a conversation about which metric would convince you.
We would rather be narrow and checkable than broad and impressive.
Every number on this site carries the method that produced it, because that is the only version of this business worth building.
Want this pointed at your fleet?
We are onboarding fleets and repair networks a few at a time. Tell us what you run and we will reach out.