AI Automated Takeoff
AI-assisted ductwork takeoff, with the estimator in control.
We're developing software that reads ductwork drawings, proposes routes, fittings and quantities, and hands them to an estimator to review, correct and accept.
Focus areas
The aim is to remove repetitive measuring, not judgement. Anything the software is unsure of is flagged for review rather than guessed, and accepted items feed the same quantity schedule as a manual takeoff.
Recognition workspace · concept
AI Review Canvas
Proposed items
Confidence Table
How review works
What it will do
- Recognition
- Identify likely duct routes, fittings, dampers and terminals across a drawing set.
- Measurement
- Propose lengths and counts by system and size for an estimator to accept or correct.
- Markup
- Generate marked-up drawings automatically, using the same conventions as manual takeoff.
- Revisions
- Compare drawing revisions and highlight what changed, so re-measurement is targeted.
Interface shown is a concept. Features and release timing will be confirmed with waitlist members first.
Design principles
Built to support estimator judgement, not replace it.
Ductwork drawings vary in quality, convention and completeness. A useful tool has to show its working and be honest about uncertainty.
Visible confidence
Each proposed item shows how confident the recognition is.
Review first
Nothing reaches the quantity schedule until an estimator accepts it.
Same output
Results use the same markup conventions and schedule layout as manual takeoff.
Planned workflow
How an AI-assisted takeoff will run.
- 01Upload the drawing set and set the scale and systems to measure
- 02Review proposed routes, fittings and terminals overlaid on each drawing
- 03Resolve flagged items — accept, correct or exclude
- 04Export the accepted quantities and marked-up drawings
Waitlist
Help shape AI takeoff.
Join the waitlist to hear about early access first. Tell us what you price and which drawings slow you down — it directly informs what we build first.