If you’ve ever driven past a construction site and thought, “That looks like chaos,” you’re not alone. But as I learned in my recent conversation with Amir Berman, VP of Industry Transformation at Buildots, there’s a whole world of precision and data science quietly operating beneath those hard hats and steel beams.
Our latest episode of Data Driven took a deep dive into the digitization of the construction industry. And let me tell you, if you thought AI was just for chatbots and marketing dashboards, think again.
Amir painted a compelling picture of the modern job site—one where computer vision and predictive analytics are transforming the way buildings come together. What struck me was how construction, an industry often perceived as slow to adopt tech, is actually a proving ground for robust, real-world AI applications.
He described it as the ultimate stress test: equipment has to be rugged, reliable, and fast. If your system goes down in the middle of a job, you’re not just losing data—you’re losing money, time, and potentially putting people at risk. There’s no room for half-baked solutions when margins are razor-thin and the stakes are skyscraper-high.
One of the key takeaways was how intuition and experience—the domain of seasoned superintendents—can be augmented by AI. Think of it like adding a digital co-pilot to a veteran pilot. Amir explained how delays that might take weeks for a human to recognize can now be flagged within days by computer vision systems tracking progress down to individual elements like ductwork or drywall.
And the data isn’t just for the here and now. Buildots leverages historical data to inform planning for future projects. It’s not just about spotting delays but preventing them before the concrete is poured or the cranes even arrive. In Amir’s words, it’s about building a factory that only gets used once, and getting it right the first time.
This isn’t sci-fi. This is today. Workers wear 360-degree cameras on their hard hats. AI systems digest the footage, compare it to 3D building plans and schedules, and produce actionable insights. It’s a bit like having Google Maps for your construction timeline, complete with rerouting when there’s a delay.
What’s even more impressive is the human angle Amir brings to this technological transformation. He talks with deep respect about the people on job sites, the pressure they face, and how AI isn’t there to replace them, but to support them. The construction industry, he points out, isn’t under-tech’d—it’s under-margined. That nuance matters. It means tech needs to be mature, reliable, and ready to go on day one.
In the end, what resonated most was this idea: Technology should serve people. It should make tough jobs more manageable, not just more measurable. And as more industries start to lean into AI, I hope they take a cue from what Amir and the team at Buildots are doing—building smarter, not just faster.
