Why vertical integration, design-for-manufacturing, and ruthless single-product focus are turning hardware founders into the next trillion-dollar companies. The industrial base didn't disappear by accident, and the founders rebuilding it are running the same operating system Elon ran twice.
"AI will be a phenomenal tool that can be used in the physical world. The reality is AI cannot turn a wrench yet." — Erin Price-Wright and Connor Love
Erin Price-Wright & Connor Love, on the episode
Erin Price-Wright and Connor Love run American Dynamism at a16z, and they spend their year visiting the mines, factories, and missile lines behind this year's AD50 list. The pop framing of this category is that it's about patriotism, drones, and steel-toed boots at remote job sites. The operating system underneath is sharper: a specific, repeatable playbook borrowed from SpaceX and Tesla, the only two American companies that have actually built physical products at scale in the last two decades, now being applied to mining, energy, and defense manufacturing. This conversation pulls the concrete mechanics out of that playbook: when to vertically integrate, how to design a product so it can actually be produced at volume, and why the founders winning right now refuse to build more than one thing until the first thing works.
Vertically Integrate Only Where No Supply Chain Exists
The play
Vertically integrate only the exquisite, no-supplier-exists components, and buy everything else from existing high-quality suppliers.
Erin Price-Wright & Connor Love says
SpaceX and Tesla vertically integrated because no industrial base existed to support them, not because it's always the right move. If you don't have a reliable industrial base to pull from, building it yourself is a hard, forced path, not a strategy of choice.
Try this
Buy from existing high-quality suppliers, and only build in-house the components with no supplier at all.
