Back to vendors
S

Struction

Also known as: Struction.co, Struction AI

Visit site
Enterprise operations agentprivateVerified 2026-07-08

AI construction estimating agent that reads a drawing set with large language models, computer vision and proprietary object detection to generate a full bill of materials, perform quantity takeoffs and price every component, with a drawing grounded chat that can modify scopes and assemblies.

Struction is an AI construction estimating tool that turns a set of construction drawings into a priced quote. It combines large language models, computer vision and proprietary object detection to read a drawing set, generate a full bill of materials, perform quantity takeoffs, and price each component and scope of work against the customer's own pricing data, claiming to estimate a whole project in about 10 minutes or produce a guaranteed maximum price in days. Its distinguishing feature is a drawing grounded AI chat: an estimator can ask why a line item is priced a certain way or what gauge studs a project needs, and Struction answers with the specific sheet and schedule it drew from, for example pointing to an interior wall schedule on a given sheet or explaining a storefront window price per square foot over the glazed area, and can then modify scopes of work and assemblies or run a new takeoff on request. Secondary sources describe the platform generating three dimensional visualized, project specific quotes by mapping extracted dimensions and specifications from drawings to product data, with an early emphasis on precast concrete work. Founded in 2024 and based in Denver, Struction was built by Joseph Rainey and Sam Tang, employs a small team, and has raised roughly $5 million. It should not be confused with Structured AI, a separate construction quality assurance and quality control vendor.

Vendor details

Canonical URL

https://struction.co/

Category

Enterprise operations agent

Funding status

Founded 2024 in Denver, Colorado by Joseph Rainey (CEO) and Sam Tang. About 10 employees. Raised roughly $5 million (reported as seed to Series A across sources; investors include Soma Capital).

Company status

private

Use cases & customers

Primary use cases

Automated construction takeoffs from drawingsFull bill of materials generation and component pricingProducing fast project quotes and guaranteed maximum priceDrawing grounded Q and A and scope adjustment via AI chat

Target customers

General contractors and subcontractors preparing bidsConstruction estimators and preconstruction teamsPrecast concrete licensors and manufacturers

Deployment options

Cloud SaaS (web platform)

Integrations

Struction ingests construction drawing sets and reads notes, labels and schedules across the set, and integrates each customer's own pricing data so quotes reflect their real costs. Its AI chat can modify scopes of work and assemblies and run new takeoffs. Broad external system integrations (for example ERP or project management tools) are not documented in retrieved sources.

Capability coverage

4.0 / 14 capabilities · 29%

Integrations & Tool CallingIngests construction drawing sets and integrates each customer's own pricing data to generate quotes, and its AI chat performs actions such as running a new takeoff and modifying scopes, though broad external system integrations are not documented in retrieved sources (struction.co/landing; PitchBook). Partial
Workflow OrchestrationRuns an automated end to end estimating pipeline, from reading drawings through object detection to generating a bill of materials, performing quantity takeoffs and pricing each component into a quote or guaranteed maximum price, and the chat can modify scopes and rerun the takeoff (struction.co/landing; Tracxn). Full
Knowledge Grounding & RAGGrounds its answers on the uploaded drawing set, reading notes, labels and schedules and citing specific sheets (for example an interior wall schedule on sheet A101), and maps extracted dimensions and specifications to product and pricing data, though grounding is scoped to the current project rather than a persistent knowledge base (struction.co/landing; PitchBook). Partial
Human Oversight & GuardrailsAn estimator reviews and refines the AI output through the chat, asking why an item is priced a certain way, modifying scopes of work and assemblies, and requesting a new takeoff, keeping a human in control of the estimate, though formal approval or confidence controls are not documented (struction.co/landing). Partial
Security, Identity & GovernanceBeyond SSL noted on a third party profile, no security certifications, role based access control, single sign on or data governance details are documented for Struction in retrieved sources (Crunchbase; struction.co/landing). Unable to verify
Observability & AuditabilityEvery quantity and price is explainable and traceable, with the chat citing the specific sheet and schedule behind a line item and showing the math (for example a price per square foot over a glazed area), though a formal audit log or observability dashboard is not documented (struction.co/landing). Partial
Memory & State PersistenceIntegrates and applies each customer's own pricing data so quotes stay aligned with their real costs across projects, giving persistent customer pricing state, though broader long term agent memory or learning from corrections is not confirmed for Struction (struction.co/landing; startup-seeker). Partial
Deployment & Data ResidencyStruction is a cloud based web platform; no self hosted, private cloud or data residency options are documented in retrieved sources (struction.co/landing; PitchBook). Unable to verify
Prebuilt Agents / Templates / PacksWorks with reusable scopes of work and assemblies that the AI chat can modify, providing prebuilt estimating building blocks, though a broader library of prebuilt agents or templates is not documented (struction.co/landing). Partial
Triggers & Channel CoverageEstimates are initiated by manually uploading a drawing set through a single web interface and chat; no automated triggers, scheduling or multichannel coverage is documented in retrieved sources (struction.co/landing). Unable to verify
Model Flexibility & RoutingStruction uses its own large language models, computer vision and proprietary object detection; no user facing model choice, routing or bring your own model capability is documented (struction.co/landing; Tracxn). Unable to verify
APIs / SDKs / MCP ExtensibilityNo public API, SDK or MCP server for external extensibility is documented for Struction in retrieved sources (struction.co/landing; PitchBook). Unable to verify
Testing, Debugging & OptimizationBeyond the core ability to rerun a takeoff, no distinct testing, debugging or evaluation tooling with accuracy metrics or confidence scoring is documented for Struction in retrieved sources (struction.co/landing). Unable to verify
Browser / Computer-useStruction analyzes uploaded construction documents; no browser control or general computer use capability is documented in retrieved sources (struction.co/landing). Unable to verify

Recent platform changes

No recent material changes tracked yet.

Pricing

No public pricing found; pricing appears to be handled directly with the company.

Not published; presumed direct enterprise pricing

Contact onlyMedium variable cost

Cost watchouts

Pricing model unknown; construction estimating tools sometimes bill per project, which can add up across many bids.

Variable cost rationale

Struction's pricing model is not published, so variable exposure is uncertain; construction estimating tools are commonly priced per seat or per project, either of which could scale with usage.

Sales call required

Yes — required for paid access

Free / trial

No free tier or trial confirmed in retrieved sources

Lowest paid plan

Not published

Key ambiguities

No pricing is published; the billing model (per seat, per project or platform license) is unknown.

Verified 2026-07-08

Contact us

Found a vendor we missed? Have feedback on the index? We'd love to hear from you.