Resource search

Search DataDrivenAEC

Type at least two characters to search.

Looking for live projects? Explore open projects or browse the feed.

Open the full search page →

SewerAI

AI computer vision for sewer pipe defect coding and condition assessment

Cloud platform (Pioneer) with an AI computer-vision engine (AutoCode) that analyzes CCTV sewer inspection footage to automatically detect and code pipe defects to NASSCO PACP/MACP/LACP standards, then turns the coded inspection data into GIS-mapped, risk-scored capital rehabilitation plans (Risk & Rehab).

Pricing
free tier, AutoCode is priced per linear foot at $0.30/ft
Alternatives
Birdsview, Fleetio, Boston Dynamics
Who it's for
Civil Engineering, Construction, Facility Management
$25M total, including a $15M Series B (announced June 2024, led by Innovius Capital, with Emerald VC, Epic Ventures, Suffolk Technologies, Bentley Systems, and Burnt Island Ventures participating) raisedGrowth Stage
Founded 2019 · Vendor states it serves customers "from mid-size municipalities to the largest water systems in the US and Canada." Named references include the City of Houston (largest US sewer collection utility, cited 55% reduction in contractor data-submittal failures and $1M+ cumulative savings under an EPA consent decree), City of Phoenix (selected top performer among 5 AI vendors evaluated with engineering firm Dibble for a 107-mile assessment program), and Sonoma County Water (97% average coding accuracy). Reported at 105-111 employees (Tracxn/PitchBook, 2026). Exact total customer count not publicly disclosed.
Website → ← All Tools
WHY IT EXISTS

Why This Tool Exists

The Problem

Municipal sewer crews and contractors manually review CCTV inspection footage frame-by-frame to catalog pipe defects against NASSCO's 226 PACP codes -- a slow, inconsistent process with a documented error rate around 20% -- while utilities under EPA consent decrees (e.g. the City of Houston, the largest US sewer collection utility) face deadlines to assess and rehabilitate thousands of miles of aging pipe.

The Solution

AutoCode's computer-vision models analyze CCTV inspection footage and manhole video, automatically detecting and coding defects (fractures, corrosion, root intrusion, blockages) to NASSCO PACP/MACP/LACP standards at a vendor-claimed 6x the speed of manual review. Coded data flows into the Pioneer cloud platform for review, GIS mapping (Esri ArcGIS sync), and Risk & Rehab capital-planning, with Sewer3D generating 3D manhole models and Smart Project Builder converting inspection/GIS data into prioritized rehab project plans.

USAGE

How You Use It

Delivery Method
SaaSCloud PlatformEnterprise API
Integrations
Esri ArcGISCityworks (Trimble)Cartegraph/OpenGovUnearth (Procore)LucityInforDeep TrekkerAries IndustriesTREKK Design Group (TREKK360 camera)Edge AI Solutions (PipeDream Robot-as-a-Service)
Disciplines
Civil Engineering, Construction, Facility Management
Workflows
Data & Automation, Operations & Facility Management
Project Phases
Operations/FM
Project Types
Infrastructure
TRANSPARENCY

Data Transparency

Exactly what this tool uses and how

Input
What it needs
Required:CCTV sewer/pipe inspection video footage
Optional:Manhole inspection video/images (for Sewer3D), Existing MACP inspection data, GIS parcel/asset data (Esri), Work-order data from connected CMMS
Formats:CCTV inspection video (specific codec/container not publicly specified), NASSCO PACP/MACP/LACP data exports, GeoJSON/shapefile (manual GIS upload)
Output
What you get
Format:NASSCO PACP/MACP/LACP-coded inspection data delivered via the Pioneer cloud dashboard
Fields:Coded pipe defects (fractures/cracks, corrosion/deterioration, blockages/debris, root intrusion), Defect severity and risk scores, Asset condition ratings, Rehabilitation/capital-plan recommendations (Risk & Rehab), 3D manhole models and point clouds (Sewer3D)
Algorithm
How it works
Model:Proprietary deep-learning computer-vision models trained on SewerAI's own labeled inspection-video library (specific model architecture not publicly disclosed)
Accuracy:Vendor markets "99%+ accurate" coded data on its main site, while a separate SewerAI resource states accuracy "ranging from 95 to 100 percent" -- the two figures are not reconciled on a single page. In a City of Phoenix/Dibble Engineering evaluation of five AI vendors, SewerAI was selected as top performer for accuracy and cost per linear foot; a Sonoma County Water case study reports 97% average coding accuracy. No independently audited accuracy benchmark against certified human PACP coders was found.
Privacy
How your data is protected
Retention:Not publicly specified for customer inspection video/coded data. The public privacy policy addresses only website-visitor data (IP address, browser/session info, cookies) and does not state a retention period for customer inspection footage or coded output.
Training:Not publicly specified whether individual customer inspection footage is used to train AutoCode models; the company states its training library has grown to 135+ million feet of processed pipe footage over time.
Compliance:Vendor states SOC 2 Type II and ISO 27001:2022 certification via its Trust Center (trust.sewerai.com) and an "AICPA SOC" badge on the site footer -- trust.sewerai.com returned an access error on direct fetch during this research, so the certification claim was not independently verified against the live page.
API
Integration
Endpoint:Enterprise API available for extracting Pioneer data (no public endpoint/documentation found); contact vendor for API access
Method:Not publicly documented
USE CASES

Use Cases

  • ·Automating NASSCO PACP/MACP/LACP defect coding from CCTV sewer inspection footage to speed up condition assessment
  • ·Meeting EPA consent-decree inspection and reporting deadlines for large sewer collection systems (e.g. City of Houston)
  • ·Generating risk-scored capital-improvement and rehabilitation plans from existing CCTV/MACP inspection data (Risk & Rehab)
  • ·Producing 3D manhole models and point clouds from manhole inspection footage (Sewer3D)
  • ·Syncing coded inspection data with GIS (Esri ArcGIS) and CMMS/work-order systems (Cityworks, Lucity, Cartegraph/OpenGov) for asset-management workflows
PRICING

Pricing

Free
No public self-serve free tier. A "Free Preliminary Rehab Plan" ($0) offers top-risk-asset identification, structural benchmarks, and cost estimates from a customer's existing inspection data. A "3x3 Pilot Offer" provides AI-powered inspection, reporting, and full Pioneer access for 3 months at $14K, cancelable anytime.
Pro
AutoCode is priced per linear foot at $0.30/ft (vendor comparison: typical manual contractor coding runs $1-$3/ft), with volume-based pay-as-you-go discounts applied automatically and additional discounts for annual commitments. Pioneer is sold as an annual, seat-based subscription (per-seat price not publicly disclosed).
Enterprise
Risk & Rehab is priced on a population (asset-count) basis; custom/enterprise pricing available via sales -- vendor asks about inspection volume, workflows, and goals to scope a package.
Sources & Research NotesResearch date: 2026-07-17
Data Sources (15)
Fields Checked (16)
problem, solution, deliveryMethod, integrations, disciplines, projectPhases, projectTypes, yearFounded, customerBase (named case studies), fundingTotal (Series B amount, lead investor, total), input-formats, output-fields, algorithm-approach, pricing-free, pricing-pro, useCases
Not Found (8)
exact data-retention period for customer inspection footage/coded data, whether individual customer footage is used for model training, specific hosting provider/data-center region, independently audited accuracy benchmark, SOC 2/ISO 27001 certification confirmed on a directly-accessible primary source (trust center returned an access error), public API documentation (endpoint, auth, rate limits), exact total customer count, per-seat Pioneer subscription price
About this research: This record summarizes the cited sources and fields checked. A research date does not establish hands-on testing. Consult the original sources for current details and limitations.
ALTERNATIVES

Similar Tools

📡
Birdsview
Data & Automation

Handheld scanner combining ground-penetrating radar, LiDAR, and camera sensors with on-device AI to map rebar location, cover depth, and diameter inside concrete structures within minutes, without drilling or facility shutdowns. A related product line, Birdsview Wood, applies the same radar-plus-AI approach to detect internal rot in wooden utility poles.

View Details →
🚛
Fleetio
Data & Automation

Cloud-based fleet management software that automates maintenance scheduling, tracks compliance, manages work orders, and provides mobile access for construction teams to optimize fleet operations.

View Details →
🤖
Boston Dynamics
Data & Automation

Deployer of Spot (4-legged mobile robot), Stretch (case handling), and Orbit (data platform). Spot provides autonomous construction site documentation, digital twin creation via reality capture (3D point clouds, 360° images, thermal imaging), and hazard detection without human site entry. SDK-enabled for custom payload integration and autonomous mission execution.

View Details →