Datagrid
AI agents that execute construction workflows end to end, from document search to submittal reviews and field observations
datagrid.comProduct Overview
What It Is
Datagrid deploys specialized AI agents that execute construction project workflows autonomously: document search, submittal reviews, RFI creation, field observations, daily reporting and estimation. Its multimodal reasoning agent analyzes drawings with dimensional accuracy, cross-references RFIs against specifications, and runs multi-step tasks that previously took multiple team members across multiple systems.
The Problem It Solves
Current Approach
Construction project teams spend hours on administrative tasks that require cross-referencing multiple document types, creating RFIs, reviewing submittals, analyzing field videos and filling out forms. Project managers and superintendents manually search through drawings, specifications, RFIs and shop drawings to answer questions or verify information. A single submittal review can take days or weeks, and field observations from video require manual transcription and documentation. These processes are time-consuming, error-prone, and pull people away from higher-value decision-making.
How It Changes
Instead of spending 20 days reviewing 20 submittals, teams complete the same work in a couple of days. Field superintendents can text a phone number or drop a video into the system, and the AI extracts issues, creates observations and drafts RFIs automatically. The multimodal reasoning agent analyzes drawings with dimensional accuracy, cross-references RFIs against specifications, and executes multi-step tasks that previously required multiple team members and systems.
Value Proposition
Dramatic time savings, with a 10x improvement on submittal reviews, 20 days reduced to 2 for 20 submittals. Reduced administrative burden on project managers, project engineers and superintendents. Improved accuracy through AI that cross-references every project document. Field-friendly access over SMS, requiring no behavior change. And a humans-in-the-loop approach, so engineers certify all decisions.
Area of Focus
Datagrid targets general contractors on active construction projects, addressing project execution workflows including document management, submittal reviews, RFI creation and management, field observations, daily reporting and cost estimation. It is designed for mid-to-large general contractors with $500M to $10B in annual revenue, managing complex projects with extensive documentation requirements.
Key Features
Deep Search Agent
Pain: Finding specific information across thousands of drawings, specifications, RFIs and shop drawings is time-consuming and often incomplete. Critical details such as dimensions, materials or installation requirements may be scattered across multiple document types, and specifications can be overridden by RFIs that are not always reflected back in the specs.
The Deep Search agent uses a specialized model that analyzes drawings with precision, understanding dimensions, locations, materials and installation details. It compares documents against each other and cross-references RFIs that override specifications.
Super Assist Agent
Pain: Field teams record hours of video and audio during walkthroughs, but turning that raw footage into actionable observations or RFIs takes significant manual effort. Quality is often poor, with shaky cameras, background noise, and several issues captured in a single recording.
The Super Assist agent performs multimodal analysis on field video and audio, extracting observations, issues and next steps automatically. It processes audio, visual content and even text visible in the video, such as drawings shown on an iPad, at the same time.
Project Management Agent
Pain: Project managers and project engineers spend enormous time on submittal reviews, RFI responses and scope of work documentation. A typical submittal review means cross-referencing specifications, drawings and standards, which can take days per submittal, and creating a detailed scope of work requires synthesizing information from multiple sources.
The PM agent executes project management workflows including submittal reviews, RFI management and scope of work creation. It analyzes complex submittals against all project documentation and company standards, identifying compliance issues and generating detailed review comments.
Estimation Agent
Pain: Creating accurate estimates means comparing current project requirements against historical data, but that information is often siloed or inaccessible.
The Estimation agent combines historical project data with current estimates to evaluate accuracy and identify potential issues, analyzing the current estimate against past performance to flag items that may be over or under-estimated.
Used By
Target Customer Types
MJ Harris
$100M revenue, 121 employees
“After seeing all the AI tools out there, Datagrid stood out as the one that was easy to use and to trust. We have it on two job sites and superintendents can find answers to questions in seconds instead of hours.”
—
Provizual
$8M revenue, 10 employees
“We build agents that navigate the web and collect more than 2,000 permits and city inspections every day. Datagrid powers our permitting data and accelerates time to value for our customers.”
—
Operating Locations
Throughout the United States, and expanding into Europe with several European customers now onboarded.
Implementation & Onboarding
Datagrid emphasizes rapid deployment. Self-service onboarding is available, so customers can create accounts and start connecting data immediately, setting up direct data connections themselves. Agent creation is accessible to users without technical expertise. Standard deployments using the five core agents launch the same day. Project-by-project implementations take two to three days. Custom use cases take roughly two weeks.
Security & Privacy
Integrations
ROI & Value
Demonstrated ROI
A 10x improvement on submittal reviews, completing 20 submittals in 2 days rather than 20. Administrative burden drops as hours-long tasks become minutes. Accuracy improves through AI that cross-references every project document to catch overrides and conflicts, and field efficiency increases through instant answers over SMS.
Success measurement
The enterprise plan provides usage analytics showing which agents are used, by whom and for what tasks, so companies can track engagement, identify power users and measure productivity improvements.
Founding Team
Thiago Da Costa
Founder and CEO
An AI product leader and entrepreneur focused on building agentic systems for real-world workflows. Thiago is the founder and CEO of Datagrid, which was acquired by Procore Technologies, where he now leads Data and AI, driving autonomous AI solutions for the construction industry.
LinkedInFunding
Stage
Acquired
Acquired by Procore Technologies