Best of LinkedIn: Digital Construction CW 34/ 35

Construction's AI conversation is splitting into two tracks that rarely meet. One track is strategic: leadership, data readiness and culture are named more often than the model itself as the real constraint on adoption. The other is operational: reality capture, BIM interoperability, generative scheduling and robotics keep compounding, each pushing more structured data into systems that AI increasingly depends on. Funding and M&A are following that second track, concentrating around whoever owns the workflow and the data behind it, while the career ladder for the people who build all of this keeps stretching further into strategy and leadership.

Date

August 31, 2026

Digital Construction

Thomas Allgeyer

Methodology: Every two weeks we collect most relevant posts on LinkedIn for selected topics and create an overall summary only based on these posts. If you´re interested in the single posts behind, you can find them here: https://linktr.ee/thomasallgeyer. Have a great read!

Listen to our podcast

If you prefer listening, check out our podcast summarizing the most relevant insights from Digital Construction CW 34/ 35

AI Adoption and Data Readiness

AI rollouts fail when leadership gives one message to both the camp eager to adopt and the camp afraid of losing their job
The fix sequences as data architecture first, then platforms, then AI, with RICS research naming skilled people (46 percent), system integration (37 percent) and data quality (30 percent) as the leading adoption barriers
Construction's AI gap reads as a culture problem, not a technology one, given fragmented systems and inconsistent reporting
Digital adoption in construction has reached 85 percent while productivity growth has stalled, a transformation paradox
The new RICS AI report finds the industry has now crossed the inflection point the same report flagged a year earlier
Failed AI pilots trace to unclear prompting on complex processes rather than to the model or the industry's readiness
Field adoption ties to voice and multimodal capture that fits into work superintendents are already doing, not a new app to learn

‍

AI Agents in Practice

Procore's release of 20 specialized AI agents includes one used by Haskell to cut submittal review from seven days to ten minutes
A catalog of 16 Claude skills spans drawing takeoffs, spec compliance, BIM model health, cost auditing and federal opportunity tracking
IFC based MCP servers offer a lighter route to BIM agents than embedding logic inside Revit itself
Model Context Protocol paired with Autodesk Navisworks can automate clash detection and coordination insight
AI speeds up information processing but does not replace the field judgment a superintendent still has to exercise on site

‍

Data Architecture and Trust for AI

Structured retrieval completes 95 percent of answers on cross referenced project documents against 25 percent for standard retrieval, with an 845 million dollar data acquisition this month as evidence structured data is becoming the moat
Project teams lose over 14 hours a week to non optimal activity hunting for data already structured at the source, and a Stanford finding shows working from the model instead of the drawing cuts estimating time by 80 percent
Enterprise AI adoption is blocked by memory, making the case for a construction knowledge graph as project memory rather than another point solution
A shared construction ontology paired with an evidence layer connects AI answers back to the drawings, models and records behind them
A single data fusion engine is needed to reconcile BIM, GIS, LiDAR, IoT and ERP data that current API to API integration cannot resolve
AEC's next opportunity is connecting engineering expertise, data and automation rather than adopting more standalone tools
People need training on the foundational "why" behind processes at the same rate AI models are being trained on them

‍

Reality Capture and Scan to BIM

The Leica RTC500 delivers 1.5 millimeter 3D point accuracy at 10 meters and a 72 megapixel six camera system capturing 432 megapixels of HDR data in about 30 seconds
In house terrestrial laser scanning prices at 80,000 to 120,000 dollars in equipment alone before software, training and utilization are factored in
A handheld SLAM scanner can capture a residential building's structural and architectural detail as a 3D point cloud for as built documentation
Reality capture in NUBIGON can simulate real world sunlight on a 300 year old church without building a conventional model first
A four step workflow turns a sample point cloud into a Revit model at a client specified level of development from 100 to 500
High precision laser scanning connected to BIM modelling links directly to fewer clashes, less rework and fewer site delays

‍

BIM Coordination and Digital Twins

BIM is the digital product definition, while VDC is how that model gets coordinated, sequenced and delivered on site
Architectural, structural and MEP models can be coordinated into one buildable design before it reaches site
Testing on a real office project shows a one to one BIM data exchange between Archicad and Revit without lost object intelligence
Enterprise BIM strategy asks how a BIM department evolves from project support into a strategic function
BIM as connected building data can progress toward machine learning that detects deviations and risk earlier
Digital twins must model power, connectivity and fallback alongside the asset itself, since a healthy sensor status can still hide an invisible failure

‍

Robotics, Automation and Additive Construction

Three autonomous excavator startups, Bedrock Robotics, TerraFirma and Gravis Robotics, have raised a combined 570 million dollars in six months
Fully driverless excavators are now running on three named US jobsites, including a 1.2 million cubic yard sitework project with Zachry Construction Corporation
Robotics ROI proves out by validating one application on one project before scaling to multiple robots and multiplying the measured productivity gain
A 3D printed concrete motorcycle shelter for ZÜBLIN in Stuttgart shows the architectural freedom offsite printing unlocks
A digital twin connected directly to a 3 ton robotic concrete printer feeds live sensor data back for predictive quality checks during the print

‍

ConTech Market, Funding and Platforms

US AEC software raised 616 million dollars across 46 rounds in H1 2026, almost double all of 2025, alongside acquisitions including Procore buying Datagrid AI, Autodesk buying Rhumbix and Nemetschek Group buying HCSS
The connected construction equipment market is sized at 4.61 billion dollars in 2025, growing to 11.57 billion dollars by 2034 at a 10.76 percent CAGR
The FHWA has opened 34 million dollars in funding for advanced digital construction management systems, with applications closing September 16, 2026
Togal.AI's Assemblies launch, starting with drywall estimating, signals AI value shifting from horizontal platforms toward trade specific workflows
A new AI agent in Constructable built a submittal log directly from drawings and fixed 26 of 44 subcontracts still sitting at 10 percent retention
Microsoft Project Online's retirement on September 30, 2026 is the trigger to evaluate scheduling alternatives that integrate cost, BIM and actuals

‍

Project Delivery, Scheduling and Controls

Generative scheduling engines can simulate over 10,000 schedule combinations against constraints like crane access, crew velocity and supply chain lead times
A five step discipline, define, document, price, schedule, approve, stops small changes from becoming unmanaged scope creep
Lean construction structures as a pyramid, with philosophy and principles at the base, tools like Last Planner and Takt Planning in the middle, and results at the top
A 212 million dollar Progressive Design Build RFP in San Diego traces back to years of positioning before it was ever published
The KREODx AEC Project Assembly User Guide connects design, DfMA, procurement, manufacturing and handover into one controlled structure
Accumulated ConTech stacks draw scrutiny from opposite angles, one account recalling a business paying for three separate PDF viewers, another asking whether anyone revisits whether the whole stack still makes sense together

‍

Revit Tooling and Firm Built Automation

Outsourced structural steel detailing on Tekla and SDS2 cuts internal drafting overhead by up to 30 percent while adding automated connection design to AS 4100 and AS/NZS 5131
Built in Revit features remove everyday friction, while custom plugins scale that productivity across an entire team
A managed deployment and maintenance lifecycle for firm built Revit add ins addresses the backlog risk of stacking internal tools without support
Revit models, Navisworks clash reports, model health reports, MIDP and RFIs connect into one dashboard through a custom add in tied to ACC
An MTO dashboard keeps material quantities synchronized with the latest BIM model through the CDE, removing manual recalculation after every design change
Autodesk Forma's July update adds AI powered room detection in beta, reusable search sets across data management and model coordination, and support for up to 2,000 values per custom attribute

‍

Thanks toAJ Waters, Ali Taheri, Arun Aravindan, Austin Lay, Bert Azizoglu, Charlie Elton, Colin Murtagh, Derek Burgin, Doanh Do, Erdem Evren, Hamza Abdul Jabbar, John Harding, John Lehmann, Joseph Trujillo, Joshua Kwaku Botozah, Kieran Lean IEng, Kristijan Vilibić, Luda Safransky, Manash Chatterjee, Oksana Kozak, Prof. Dr. Lisa Lenz, Roman H., Rusbe Kal ‍, Kristopher Lengieza, Brett King, Guido Maciocci, Adesuwa Aghedo, Angelo Kirchon, Chris Nisbet, Chun Qing Li, Francisco Maranchello, James Gray, Kelvin Okolieaboh, Mahmoud A. Hamed, Molly (Stites) Abbott, Saad Hussein, Stephanie Brown, Tomislav Žigo, Alief Muhammad Farhan, Anil Sawhney, Benjamin TAN, Simran Kanwar, Vladimir Vlahovic and everyone else who contributed insights to this edition. linkedin.com/pulse/best-linkedin-cw-34-35-digital-construction-thomas-allgeyer-epbwf/?lipi=urn%3Ali%3Apage%3Ad_flagship3_profile_view_base%3BuMMfPXD5TD2jzMjQ3rX7NA%3D%3D