Work & Research

Real projects. Measurable results.

From enterprise digital twins to published research — a selection of projects that demonstrate what AI-powered building intelligence can do.

AllConstructionGovernmentEnergyResearch

5

Projects

2

Client Deployments

2

Publications

Client ProjectConstruction · 2025

Enterprise-Scale Digital Twin Platform

Fortune 500 Construction Company

Challenge

A Fortune 500 construction company needed real-time spatial analytics and digital twin capabilities across large-scale projects. Existing tools couldn't handle the data volume or complexity.

Solution

Built a scalable digital twin platform integrating IFC processing, spatial analytics, and real-time IoT data streams with a custom WebGL visualization layer.

  • Scalable IFC processing pipeline
  • Real-time spatial analytics engine
  • Custom WebGL visualization layer
  • Integration with IoT sensors and BMS

The digital twin platform has revolutionized how we manage and analyze our construction data.

VP of Innovation, Construction Technologies

Digital TwinIFCSpatial AnalyticsIoT

2.5M m²

Building Space Analyzed

100+

Concurrent Users

40% faster

Clash Detection Speed

300%

ROI in Year 1

IFCDigital TwinReal-time · AI · IoT
Client ProjectGovernment · 2024

Government BIM Validation Platform

Government Infrastructure Division

Challenge

Manual BIM compliance validation for public infrastructure projects was taking weeks per submission and producing inconsistent results across reviewers.

Solution

Deployed an automated IFC validation platform using custom rule engines, spatial graph analytics, and natural language query interface for non-technical reviewers.

  • Custom IFC validation rules using IfcOpenShell
  • Spatial graph analytics with Neo4j
  • Azure OpenAI integration for NL queries
  • Automated reporting and visualization dashboard

Intento Labs transformed our BIM validation process from weeks to hours while significantly improving accuracy.

Senior Director, Digital Infrastructure Division

IFCValidationComplianceGraphRAGGovernment

Weeks → Hours

Validation Time

88%

Accuracy

5× faster

Reviewer Efficiency

NBC + OBC

Standards Covered

Egress WidthFire DoorsAccessibilitySmoke DetectorsStair Ratio80%Compliance
ResearchResearch · 2023 Published Research

Automated Space-Based Graph Generation for Building Energy Estimation

Research Publication

Challenge

Employing BIM data for Building Energy Consumption Estimation (BECE) is challenging due to complex IFC data models and incompatibility with data-driven ML algorithms.

Solution

A framework that extracts semantic, geometry, and topology information from IFC space schemas using geo-computation algorithms, making the data compatible with graph-based machine learning.

  • IFC space schema parsing and enrichment
  • Geo-computation for adjacency detection
  • Graph construction from building topology
  • Graph-based ML for energy estimation
BIMGraph MLEnergyIFCResearch
Read the paper

90%+

Geometry Accuracy

100%

Adjacency Detection

Graph-ML Ready

Data Format

MDPI Buildings

Published

JFMAMJJASONDBaselineOptimisedkWh/m² · EPW Climate Data
Coming SoonComing SoonEnergy · 2026

IFC to IDF Energy Processing & Carbon Footprint Engine

In Development

Challenge

Building energy simulation requires converting rich IFC models into EnergyPlus IDF format — a process that today is largely manual, error-prone, and disconnected from real-world climate data. Carbon footprint calculation requires integrating EPW weather files with the building's thermal model, but no automated end-to-end pipeline exists.

Solution

An AI-powered pipeline that parses IFC models, extracts thermal zones, materials, and HVAC systems, auto-generates IDF simulation files, and ingests EPW climate data to compute per-zone energy consumption and operational carbon footprint — all in one workflow.

  • Automated IFC → IDF conversion with IfcOpenShell + EnergyPlus
  • EPW weather file parsing and climate zone mapping
  • Thermal zone extraction from IFC space schemas
  • AI-assisted material property mapping to EnergyPlus datasets
  • Per-zone carbon intensity calculation (kg CO₂e/m²)
  • Scenario comparison: baseline vs. retrofit vs. passive design
IFCIDFEnergyPlusEPWCarbonClimateEnergy Simulation

Hours → Minutes

Conversion Time

EPW Global

Climate Datasets

kg CO₂e/m²

Carbon Metrics

EnergyPlus

Simulation Engine

BuildingFloor 1Floor 2Space ASpace BHVACStruct
ResearchResearch · 2022 Published Research

Graph-Based Learning Using BIM for Energy Consumption Estimation

Research Publication

Challenge

BIM data is severely underutilized for data-driven energy approaches due to the complexity of IFC formats and their incompatibility with standard ML pipelines.

Solution

Demonstrated graph-based learning algorithms using enriched semantic, geometry, and topology information extracted from BIM for critical zone detection in energy analysis.

  • BIM semantic enrichment pipeline
  • Graph neural network architecture
  • Critical zone detection model
  • Pre/post-construction performance comparison
Graph MLBIMEnergyNeural NetworksResearch
Read the paper

Validated

Critical Zone Detection

Graph-ML ready

BIM Compatibility

Pre + Post Build

Scope

TCRC 2022

Published

BuildingFloor 1Floor 2Space ASpace BHVACStruct

Ready to be the next case study?

Tell us about your project. We'd love to understand your data challenges and explore what's possible.