reducing risk for real projects in the real world

Owner, Transit Agency

[updated 3/2025]

Sound Transit (ST), the metro Seattle public transit authority, is overseeing the nation's most ambition transit expansion at over $100b and 120 miles over the next three decades. With as-built conditions needing approval before going into operation, ST's Design Technology team of five is tasked with providing a safety confidence score to the Chief Engineer for critical "go, no go" decisions for opening each section for public use. Reviewing nearly ~4500 sheets of as-built drawings biweekly, this team turned to TwinKnowledge to drastically increase review efficiency and thus confidence levels for the Chief Engineer.

Working closely with the Design Technology team, TwinKnowledge identified a few key opportunities in their current workflow for AI to drastically increase their efficiency: 

• for each review item, information has to be found within and then mapped between three different sources -- drawings, pdf documents, and spreadsheets --, some thousands of pages long
• these documents are scattered and have to be found and then retrieved from different places across the project's Sharepoint: ST's hub, contractor's hubs, and sub-contractor's hubs
• each contractor/subcontractor records information in different ways, making it difficult to standardize review

With it's construction AI, TwinKnowledge saw an opportunity to automate the tasks behind these three issues and completely remove them from the Design Technology Team's manual process: scouring Sharepoint for the right files, finding and mapping relevant information, and standardizing the review process across inconsistent information records.

Architect

[updated 3/2025]

Having been named Fast Company’s Most Innovative Architecture Firm in the World, SHoP Architects know a thing or two about good innovation. When TwinKnowledge presented their AI Assistants to SHoP’s senior executives, they quickly recognized their immense potential to reduce the 75% of time that project managers spend searching for answers to RFIs in construction administration (CA).

Working closely with SHoP’s CA team on the Fashion Institute of Technology (FIT) expansion, TwinKnowledge quickly discovered that the majority of their RFI response pain came from three cases:

• Trying to extract exact answers – numbers, specifications, detail information – from within thousands of pages of information
• Needing answers that spanned both specifications and drawing set
• Uncovering the story of how a design component changed throughout the project

Having come onto the project in the middle of CA, these problems were exacerbated for the FIT CA team. TwinKnowledge saw an opportunity to match the construction information understanding of its TK Assistants with the design judgment expertise of SHoP’s CA team, automating the information search and retrieval costin SHoP so much time. 

Owner/Operator

[updated 3/2025]

Toll Brothers has been one of Fortune Worlds' Most Admired Companies 10 years in a row for a reason. With so many housing projects going on across the country at one time, Toll was looking to improve how they transfer learnings forward and standardize designs from project to project, creating operational efficiency and QA/QC practices at scale.

In collaboration with Toll's West Coast Director of Quality Assurance, TwinKnowledge was tasked with leveraging their AI Assistants to identify discrepancies in contract documents during the initial project phases, an area identified by Toll as the source of 75% of their project issues. Through weeks of diving into Toll's review workflows, TK discovered that this issue stemmed not only from the difficulties in matching corresponding information from two dependent contract documents, e.g. the Scope of Work and the appropriate drawing details, but the fact that Toll's sources of truth for bidding documents were scattered across two departments, multiple documents, and multiple systems of record. TK saw an opportunity to unify all contract document information within a single knowledge base, allowing corresponding information from across them all to be easily compared side-by-side to ensure alignment between all documents. In expanding their document alignment capabilities at the very beginning of projects by multiples, Toll saw the potential to eliminate up to hundreds of downstream RFIs caused by design misalignment.