Turning natural language into validated Bills of Materials — from concept to kit.
We live in a golden age of software, but hardware is still gated. If you have an idea — a solar-powered cat feeder, a custom macro-pad, a CB750 engine rebuild — you hit a wall. You don't know which parts are compatible, where to source them, or what a real BOM looks like. That gap between "I want to build this" and "here's what to order" costs hobbyists, engineers, and small shops enormous amounts of time.
BuildSheet is a generative hardware platform that turns natural language into structured, sourceable Bills of Materials.
All outputs are draft-quality and intended to accelerate the documentation process — not replace engineering judgment. Every generated BOM should be reviewed and verified before procurement or fabrication.
BuildSheet was built by Blake Barrett during the Google AI Hackathon 2025, starting from a real problem: documenting a Honda CB750 engine rebuild. Finding the right parts across a dozen vendors, cross-checking compatibility, and keeping a clean BOM was a full-time job on top of the actual wrenching.
The hackathon produced a working prototype. Since then, BuildSheet has grown into a production-grade web app with Firestore-backed projects, Stripe-powered subscriptions, and a full suite of export and collaboration tools.
BuildSheet is open source. View the code on GitHub. For questions or enterprise inquiries, reach out via GitHub Issues.