AnalyzTech

01 / Brief

Solution

AnalyzTech helps technical engineers and compliance controllers run floorplan and document verification with AI. The project started as R&D to define how to read floorplans, build datasets, and design reliable detection and classification workflows. It moved to MVP and is now in production. I built the project end to end: backend APIs, model pipeline, frontend floorplan polygon visualization, AI Agent skills for verification actions, AWS deployment, and integration across the stack. Before AnalyzTech, teams had to manually read 40+ page documents and execute compliance checks by hand on plans. The platform reduced repetitive manual work and made verification faster and more consistent.

02 / Constraints

Problem

• Building a reliable floorplan reading workflow from scratch (dataset, detection, classification, extraction)
• Improving robustness on large documents and noisy plan structures
• Reaching high accuracy for room classification and room-mask detection
• Reducing latency so users can run verifications without long waits

03 / System

Architecture / workflow

• Pipeline: PDF input -> Kreo wall/door detection -> YOLO furniture/open-door detection -> room extraction -> VLM room classification -> post-processing -> AI Agent verification
• Full FastAPI backend with integrated verification workflows and project context routing to the AI Agent
• Frontend integration to visualize polygons on plans and trigger AI-assisted controls
• Classification and detection accuracy optimization through iterative model and post-processing tuning
• Production deployment on AWS with scalable API infrastructure

04 / Evidence

Results / impact

• 80% time saved on verification workflows
• +50% productivity gain
• 99% classification precision
• 96% room-mask precision

05 / Gallery

Visual references