Every great skyline starts with a mountain of paperwork. Long before the first excavator hits the dirt, massive commercial projects live or die in “pre-construction” — a chaotic, high-stakes phase of dense codebooks, endless 2D plans, and clash detection where estimators try to predict the future using spreadsheets. An estimator missing a single clause in a local building code or slightly miscalculating a material takeoff can completely evaporate a contractor’s profit margin months down the line. Despite building the physical world around us, the construction industry’s productivity has been notoriously flat for decades, largely because the very people planning these multi-million dollar projects are suffocating under manual data extraction.
It takes builders with a profound understanding of complex data systems to look at this pre-construction chaos and recognize it as an intelligence problem waiting to be solved. That is precisely what Kanav Hasija and Tanmaya Kala are doing with Meltplan.
Founded in 2025, the Berkeley-based team is building a deeply intelligent, construction-aware AI platform that completely transforms the slowest, most vulnerable phase of any building project.
Meltplan is fundamentally trained on municipal codebooks, architectural specs, intricate drawings, and real-world project data. It functions as an expert AI code consultant that delivers instant, fully-cited answers to complex regulatory questions. Alongside this, it generates highly precise AI material takeoffs from 2D and 3D plans that estimators simply review and approve. Instead of spending weeks manually counting materials and hunting for design clashes, pre-construction teams can collaborate seamlessly in a unified environment.
The true moat here is uncompromising domain awareness paired with radical time-savings. Meltplan respects that in construction, an AI hallucination isn’t just an annoyance; it is a financial liability. By anchoring every single insight in verifiable citations and real project data, they replace the grueling, error-prone manual estimation process with deterministic accuracy. It shifts the paradigm from hoping your margins hold up, to planning with absolute, data-backed confidence.
The market? General contractors, architects, specialty trades, and project owners who desperately need to compress their pre-construction timelines and protect their budgets before the first shovel ever hits the ground.
Seeing founders head-down, tackling the most stubborn productivity bottlenecks in the largest industry in the world, is profoundly inspiring. They are giving estimators their time and contractors their profit margins back.
Let’s celebrate the builders.
✍️
#AI #ConstructionTech #ProductNation
