When you meet us, you might be surprised: two brothers from the UK, working on US construction permitting. But there's a story that makes sense.

Growing up in the UK, we saw firsthand what happens when housing supply can't keep up with demand. Young people priced out of cities. Families stuck in cramped flats. A generation locked out of homeownership. We became obsessed with a simple question: why is it so hard to build?
Then came Grenfell. Seventy-two people lost their lives because of failures in building safety. It was a stark reminder that cutting corners costs lives. The regulations exist for a reason. But we also saw how the current system—paper-based, manual, reactive—fails both builders and the public. Compliance shouldn't mean months of delays. And speed shouldn't mean risk.
When large language models started to actually work, we saw an opportunity. Not to replace human judgment, but to give humans superpowers. To make the review process faster, more consistent, and more thorough than any human could achieve alone. We knew we wanted to build something at the intersection of construction and AI.
We quickly realized that the real bottleneck wasn't technology—it was data. Building codes exist as PDFs, scanned documents, and dense legal text. To build AI that actually works for construction, we first had to make codes machine-readable. That became our mission: transform the regulatory landscape into something AI can understand, so AI can help humans build better, faster, and safer.
And here we are.