Samuel Beek learned a valuable lesson about trusting AI with high-voltage tasks when his DIY electric door opener project went spectacularly wrong. Guided by instructions from ChatGPT, he wired up the device, only to discover the chatbot's grasp of electrical engineering was, shall we say, a bit short-circuited. The resulting misallocated power surge promptly blew every fuse in his house, proving that large language models still struggle with the wet versus dry connection distinction.

This electrifying failure became the catalyst for Beek's new venture, Schematik, which is being pitched as the 'Cursor for hardware.' For the uninitiated, Cursor is an AI-powered code editor that helps software developers; Schematik aims to do the same for the physical world of circuits and components. The idea is to provide an AI assistant that can actually help you build hardware without turning your home's electrical system into a fireworks display.

And it seems the concept has caught the attention of some deep-pocketed observers. Anthropic, the AI safety-focused company behind the Claude models, reportedly wants in on the action. Their interest suggests a belief that the messy, tangible world of hardware design is the next frontier for AI assistance, moving beyond just generating text or code.

The core promise of Schematik is to prevent precisely the kind of catastrophic yet hilarious user error that befell its founder. By providing reliable, context-aware guidance for physical builds, it aims to stop people from blindly following AI-generated instructions that might, for instance, confuse a load-bearing wire with a neutral one. It's a tool born from the smoking ruins of a fuse box, aiming to bring a much-needed layer of electrical common sense to the AI hardware revolution.