Automating cost estimation in construction with AI
We help large construction firms win more multimillion-dollar bids with AI.
What do we do?
The problem
The construction industry is one of the largest in the world — but also one of the least digitized. And nowhere is this more obvious than in the way contractors prepare cost estimates and bid on projects. In LATAM, the standard workflow hasn’t changed since the 1980s: architects send PDFs of blueprints, engineers open AutoCAD, manually count elements, transfer data into Excel, and then spend days crunching numbers just to build a basic estimate.
The process is slow, expensive, error-prone, and eats up hundreds of skilled labor hours per project — before the company even knows if they’ll win the contract.
The pain is real and universal:
- Engineersare overwhelmed, often working nights and weekends to meet deadlines.
- Construction companies bleed cash on estimates that go nowhere.
- They also need to forgo some projects due to lack of bandwidth.
- Accuracy suffers, leading to underbidding (and lost money) or overbidding (and lost contracts).
- There’s no data feedback loop, so knowledge isn’t compounding.
What are we building?
We’re building an AI-powered platform to automate takeoff and cost estimation for large construction sites – think malls, hotels, civil constructions— the most painful, time-consuming bottleneck in construction bidding. In LATAM (and much of the world), this is still done like it was in the 1980s: multiple engineers manually counting items on PDF blueprints, copying numbers into Excel, and burning hundreds of hours per project just to price a bid they might not even win. It’s slow. It’s expensive. It’s broken.
Our AI replaces that entire process, cutting 98% of the time, slashing costs by up to 90%, and helping companies win more contracts with radically better accuracy and speed. In a $12 trillion industry where margins are razor-thin and every bid counts, this is a no-brainer. And by automating the pain point where construction data is created, we’re uniquely positioned to become the data infrastructure for the entire industry.
Our vision
In the next 5–10 years, construction will be completely reshaped by AI — and yet, today, the most critical task in bidding for a project, takeoff and cost estimation, is still done with PDFs, spreadsheets, and manual counting. That’s about to change — and fast.
We’re at the intersection of three massive AI-driven shifts:
- The automation of knowledge work — Engineers and estimators won’t need to spend weeks manually quantifying plans. AI will do it faster, cheaper, and with near-perfect accuracy.
- Private data becoming the most valuable training asset — In construction, the real gold is hidden: behind firewalls, in private project files, budgets, and drawings. Public LLMs can’t access it, but the software embedded at the estimation layer can.
- Decision-making shifting from instinct to intelligence — AI will eventually guide what projects to bid on, how much to price them at, and how to allocate resources — but that starts with capturing structured data from day one.
By starting with takeoff automation — the exact moment when key data is generated — we’re not just solving a huge short-term pain point. We’re generating the future intelligent data-layer for the construction companies that will win on the long-run by being able to extract value from their own data.