Roofdraft
LiDAR drafting system for automated technical surveys.
Problem Defined
"Manual drafting latency delays projects and compounds material waste."
Strategic Context
Architectural surveys are slow, dangerous, and error-prone.
Competitive Imbalance
Survey-to-draft latency creates bottlenecks and waste.
System Hypothesis
On-device LiDAR processing converts geometry into blueprints instantly.
Process Architecture
How the system was designed, tested, and refined.
DEFINE
Identify drafting bottlenecks and material waste in surveying.
- • Conducted time-motion studies on survey crews
- • Audited manual drafting errors
- • Mapped project delays
- • Focused solely on drafting speed, ignored on-site safety risks
- • The true friction is the delay between site visit and blueprint availability
- • Pivoted to a real-time on-device generation model
MAP
Visualize the LiDAR to CAD data pipeline.
- • Mapped geometry extraction logic
- • Identified structural defect detection nodes
- • Initial maps assumed perfect lighting and clean scans
- • Real-world surveys are noisy; system must be resilient to occlusion
- • Added probabilistic filling for occluded structural nodes
VALIDATE
Test on-site blueprint generation accuracy.
- • Ran side-by-side tests with traditional tools
- • Measured survey velocity vs manual drafting
- • Early prototypes lacked the precision for complex roofing angles
- • Automation must be verifiable by the operator on-site
- • Integrated immediate visual feedback for captured geometry
EXECUTE
Engineer the LiDAR processing engine.
- • On-device LiDAR pipeline
- • Blueprinting engine
- • Defect detection model
- • Over-engineered the materials estimation early on
- • Geometry precision is the foundation; the rest is downstream
- • Focused on sub-mm accuracy for point-cloud-to-mesh conversion
MEASURE
Calculate throughput and material waste reduction.
- • Survey velocity
- • Estimate accuracy
- • Defect precision
- • Calculated waste solely as material, ignored labor time loss
- • 3x throughput is the primary driver of project profitability
- • Introduced project lifecycle tracking for holistic ROI
Rule Application
How doctrine was operationalized.
Intellectual Rigor
01_INT- • Stressing assumptions on LiDAR precision
- • Mapping structural logic before building
98% defect detection accuracy achieved in blind trials
Tactical Execution
02_TAC- • Shipping on-device viewer before full automation
- • Prioritizing core geometry engine
Core survey tool deployed before cloud sync features
Human Calibration
03_HUM- • Designing for one-handed on-site operation
- • Reducing cognitive load for surveyors
Interface optimized for high-glare environments and glove usage
Machine Leverage
04_AI- • Using AI for structural defect detection
- • Automated geometry synthesis
AI eliminates manual drafting bottlenecks by auto-filling mesh data
Product Architecture
LiDAR engine, on-device generation, defect detection.
AI Leverage
Model inference for structural defect detection.
Outcomes & Learnings
Accelerated survey cycles and increased estimate accuracy.
