Digital delivery
Model checks you can see, and repeat
Before a model is handed over, someone has to check it against the project's requirements, and on a live project that happens again and again. We built two tools that make those checks visible, repeatable and easy to trust.
See the problem
Each failing element lights up in 3D, with the expected value, the actual value and the reason.
The same review, every time
Reviews are saved as versioned workflows that run the same way on every model.
AI that explains, not decides
An assistant can choose and explain a review, but the checks run as fixed steps.
The problem
BIM coordinators and engineers check models against handover requirements: required information, naming, sizes, classification. In most tools that means exporting schedules to a spreadsheet and hunting for the failing elements by hand. And a review done differently each time leaves no reliable trail.
BIM Studio: checks you can see
Open an IFC model in the browser and see it in 3D, in plan and in section. Run your own rules, and every result shows the required value, the actual value and the reason. Failing elements light up in the model.
You can also ask a question in plain language, such as which columns are too short, and get back the list of elements, ready to highlight. Everything runs in the browser, so the model never leaves your machine.
BIM-IFC Workbench: the same review, every time
Reviews such as a handover information check are saved as versioned workflows that anyone on the team can run, export or adapt. Each result gets a fixed identifier, so a finding in a report can be traced to the exact query that produced it.
An AI assistant works on top of the same services. It can choose the right review, run it and explain the result, but it does not change how the checks run.
Why it matters
Model quality is where digital delivery succeeds or fails. When checks are visible and repeatable, teams spend less time hunting for problems and more time fixing them, and the owner can see what was checked before accepting a model.
