Inspectial

AI field inspection app · Early access

The field inspection app that turns photos into approved, map-ready findings.

Capture photos, video, voice notes and GPS offline, guided by your checklist. Inspectial's AI drafts each finding with its component, severity, confidence and the evidence behind it. A qualified inspector approves every one, and the result is a structured, geolocated dataset for your GIS and asset systems, not just a PDF.

Capture offline. AI drafts the findings. A human approves. Export GeoJSON, CSV, PDF or use the API.

See an example record
  • Works offline
  • Every finding linked to its photo
  • A qualified reviewer approves
  • GeoJSON, CSV, PDF, API

Every field photo becomes a finding your reviewer can check against the evidence.

A black cast-iron fire hydrant stands in tall roadside grass. Its bonnet paint is chipped, the cap chains are rusted, a steel cable runs across the front at nozzle height, and the base flange shows patches of paint loss.
Hydrant H-220730.3119, -95.4561 · custom template
  • Bonnet · paintchipped, bare metal at flangeAI suggested
  • Nozzles · pumper capin placeHuman verified
  • Nozzles · hose capin placeHuman verified
  • Chains · conditionrusted, attachedAI suggested
  • Base flange · coatingpaint loss, corrosion (possible)AI suggested
  • Access · obstructionsteel cable across nozzlesAI suggested
  • Access · vegetationtall grass, mowing neededAI suggested
Evidence IMG_2207, IMG_2208, VN_0031 · Confidence 0.88
{
  "asset": "H-2207",
  "location": { "lat": 30.3119, "lon": -95.4561 },
  "bonnet": {
    "paint": "chipped, bare metal at flange",
    "status": "ai_suggested"
  },
  "nozzles": {
    "pumper_cap": "in place",
    "hose_cap": "in place",
    "status": "human_verified"
  },
  "chains": {
    "condition": "rusted, attached",
    "status": "ai_suggested"
  },
  "base_flange": {
    "coating": "paint loss, corrosion (possible)",
    "status": "ai_suggested"
  },
  "access": {
    "obstruction": "steel cable across nozzles",
    "vegetation": "tall grass, mowing needed",
    "status": "ai_suggested"
  },
  "evidence": ["IMG_2207", "IMG_2208", "VN_0031"],
  "confidence": 0.88
}
The numbered boxes on the photo map to the observations in the record. The hydrant here is one example: the components, observations and condition scale come from the template for your inspection type.

Checklist apps digitise the form. Someone still types every finding.

Most inspection apps replace paper with a form on a phone. That helps, but the inspector still looks at each photo, picks the defect, writes the description and rates it, field by field. The photos end up attached to the record rather than turned into it.

Across hundreds of assets, that typing and the office write-up after it are where the hours go. And what comes out is usually a PDF or a spreadsheet, so getting the data into a GIS or an asset system means re-keying it.

  • Photosdozens per asset
  • Video clipswalk-arounds, close-ups
  • Voice and text notesunstructured
  • GPS, asset IDs, last inspectionin separate places
Then, by handtyping each finding, writing the report, re-keying data into GIS

How it works

One workflow from the site visit to your GIS. AI drafts the findings; people make the call.

  1. Capture offline, guided by your checklist

    The field app lists today's assets by distance and walks the inspector through the template's evidence checklist. Every photo, video keyframe and voice note is stamped with time, GPS and asset ID. The full flow works with no signal.

  2. AI drafts the findings

    Evidence is grouped by component and turned into draft observations: what was seen, a severity on your condition scale, a confidence and the evidence IDs it came from. Where the photos can't support a finding, the AI says so instead of guessing.

  3. A qualified reviewer approves

    A reviewer checks each finding next to its photos and approves, edits or rejects it. Every value is tagged AI suggested, edited or human verified, and only approved inspections become official records.

  4. Export structured, geolocated data

    Approved inspections go out as GeoJSON, CSV, a PDF report rendered from the approved data, or through the REST API, keyed to your asset IDs and ready for ArcGIS, QGIS, PostGIS or your asset system.

One app, a template per inspection type

A new inspection type is a new template, not a new app. We set it up with you from your current forms, condition codes and report format. These packs already have their own pages; each is a starting point we adapt to your rules.

What a template defines

  • Evidence checklist
  • Components and observation types
  • Condition scale
  • AI rules and limits
  • Report format
  • Export fields and asset IDs

Drainage and stormwater

Buildings

Infrastructure

Construction

Emergencies

Default condition scale

Good
Functions as intended. Minor, isolated defects at most.
Fair
Moderate defects that don't yet affect function or safety. Monitor or schedule maintenance.
Poor
Significant defects that affect function or shorten service life. Repair needed.
Critical
Failed, unsafe or at risk of failure. Immediate action.

Templates start from

  • AASHTO culvert and bridge guides
  • EPA stormwater permits
  • ASTM E2018
  • HUD NSPIRE
  • Your own forms and codes

Capture in the field. Review from any desk.

Field app preview: a map of today's route with markers by inspection status, working offline, with GR-12 selected and ready to inspect. Web app preview: the Spring asset survey project on a map, with geotagged field photos pinned at each site and a sidebar listing each site's review status.

Field app (iOS and Android) and review web app. In development.

AI creates the inspection record, not just the report

Keep your GIS and asset system. Inspectial automates the inspection-coding layer in front of it, under rules you can trust.

Agent-ready data, not documents
The structured dataset is the source of truth, so your GIS, scripts and AI agents read it directly. Reports are generated from approved observations, never written straight from the media.
Evidence first
Every observation references the photos, clips or notes behind it. A finding never exists only as an AI sentence.
A human approves
AI proposes; a qualified person decides. Inspectial is not an autonomous inspector, and only approved inspections become official records. Every value is tagged with who stands behind it.
Location is first-class
Evidence, observations and assets carry coordinates, so the output drops onto a map without cleanup.
Offline-first
The full capture flow works with no signal. Data syncs when the connection returns.
Configurable, not custom-built
Your checklist, condition codes and report format live in a template we set up with you. Not a new app.

Who it's for

Inspection and engineering firmsYour inspectors already shoot dozens of photos per asset. Inspectial drafts the findings from them, so the team spends its time on judgment and review instead of typing and report writing.
Utilities and municipalitiesGet condition data for hydrants, outfalls, culverts, poles or service lines that lands in your GIS, keyed to your asset IDs, without anyone re-keying it from PDFs.
Teams with photo-heavy, location-bound inspectionsIf the job is visiting assets on a map, photographing them and rating their condition, a template can capture it, whatever the asset type.

Outputs and integrations

Inspectial feeds the tools you already run. It doesn't ask you to move your asset data anywhere, and because the output is structured, evidence-linked data rather than a document, it also works as input for AI agents and automations.

Works with

  • ArcGIS
  • QGIS
  • PostGIS
  • MapLibre

Formats

  • GeoJSON
  • CSV
  • PDF
  • REST API
  • Excelplanned
  • Shapefileplanned
  • GeoPackageplanned

Questions

Something else? Write to pilot@inspectial.com.

How is this different from a checklist or form app?

A form app digitises the checklist, and the inspector still enters every finding by hand. In Inspectial the AI drafts the findings from the photos, video and voice notes, each tied to its evidence, and a person approves them. The output is a structured, geolocated dataset that GIS, asset systems and AI agents can read, with the PDF report rendered from it.

Which inspections does it handle?

Photo-heavy, location-bound ones. Our pilot is culverts, and packs with their own pages include SWPPP, MS4 stormwater, lead service lines, condition assessments, NSPIRE, facades, damage assessment, pre-construction surveys, bridges, utility poles, roofs and construction QA. If yours isn't listed, tell us what you inspect.

Can it use our own forms and condition codes?

Yes. The checklist, observation types, condition scale, AI rules and report layout are configuration in a template, not code. We build the template from your current forms and codes, and exports keep your asset IDs.

Who sets up the template?

We do, with you. There's no self-serve template editor yet: you send your forms, rating manual and a few past inspections, and we turn them into a template and check the AI's output against your inspectors' findings.

Does the AI replace our inspectors?

No. AI drafts the observations; a qualified inspector or engineer approves, edits or rejects each one. Only approved inspections become official records, and every value shows whether a person verified it.

Does the field app work without signal?

Yes. The full capture flow, including the checklist, photos, video and voice notes, works offline. Data syncs when the connection returns. The iOS and Android app is in development.

Does it work with ArcGIS or QGIS?

Through open formats. Approved inspections export as GeoJSON or CSV with coordinates and your asset IDs, or through the REST API, which ArcGIS, QGIS and PostGIS read directly. Excel, Shapefile and GeoPackage are planned. There are no native connectors yet.

What does it cost, and how do we get access?

Inspectial is pre-launch, so there's no public price list yet. Request early access and tell us what you inspect and how much; we'll set up a template with you and agree pricing before you use it on live work.

Request early access to Inspectial

Inspectial is in early access while our culvert pilot runs. Tell us what you inspect, roughly how many inspections a year, and which GIS or asset system the data should end up in. If we can build a template for it, we'll set one up with you.

  • Bring real inspections: photos, video and the reports you wrote from them.
  • We set up a template that matches your checklist, condition codes and report.
  • You review the AI's draft findings against your inspectors' own.
  • Approved data arrives as GeoJSON, CSV, PDF or through the API.