selected work

A company and five research projects, two of them published, all after the same thing: making a building tell you what is going on inside it, early enough to act on.
Twynable platform: the digital twin interface

Twynable: Google Maps for a building

Founded an AI platform that gives maintenance teams a live 3D copy of their building. Won first place at the NYUAD incubator.

Read the case
At the EC³ and CIB W78 conference in Porto, where the paper was published

Predicting indoor particle pollution three days ahead

Built a deep-learning model that forecasts indoor PM2.5, the particles fine enough to reach the bloodstream, 72 hours ahead of time.

Read the case
The scanning robot on the construction site

A robot that scans a construction site on its own

Built a mobile robot that maps an active construction site by itself, then stops at chosen points to take survey-grade laser scans.

Read the case

An air quality assistant that cites its sources

Built an AI assistant that answers air quality questions straight from the standards, citing the source behind every claim as it writes.

Read the case
Six-hour-ahead CO₂ forecast against what actually happened

Catching stuffy air six hours before people feel it

Forecasting the CO₂ build-up that makes a room stuffy, six hours ahead, tested across 11 university labs and their sensor history.

Read the case

Ten campus spaces you can walk through in 3D

Captured ten real spaces with a 360° camera and rebuilt them as 3D Gaussian splats: photorealistic models you can move through.

Read the case

If any of this overlaps with what you are building, say hello.

Get in touch