Let's build ships in America again.

A shipyard partner sent us a hull block model for one of the ships they’re building. Very cool, high-profile project. We were beyond excited: there were around 150 panels in a single block we can build robotically today, a meaningful share of the entire structure representing ~1000 hours of labor. And that’s one of 9 blocks. But this is just the beginning. Over the next 12 months our roadmap should get us to 90% coverage on hull blocks like this - and that’s before even getting into the superstructure. In practice, this will mean dramatically higher shipyard throughput. All the labor that goes into building hundreds of panels today - all slightly different, because this design has little parallel mid-body - can be redirected to more complex fabrication/assembly/outfitting tasks. Same shipyard facilities, same labor force, more tonnage on the water. We are going to accelerate so many ships.

If you’re going through hell… keep going.

In awe of SpaceX and its story - past, present and the future. You can think about it in 10 different ways and continue re-blowing your mind in circles. Huge congrats to the team! 🚀

The politics of envy is so caustic. America is the only place in the world where you can start from nothing, pioneer insane technology that was the stuff of Sci-Fi until recently, and *create* unimaginable wealth that never would have existed otherwise. Zero sum is a lie.

Today, we're starting with simple things: robotic cells that make hull-building faster. In the future, we will be the driving force behind a new breed of U.S.-built commercial ship: nuclear-electric (powered by the likes of @AaloAtomics, @valaratomics, etc.), minimally-crewed, and primed to take market share back from China. But we'll do it by enabling American shipyards, not competing with them. We'll create or adapt designs for manufacturability, our robotics platform will provide scale, and shipyard partners across the country will provide assembly & outfitting capabilities. This is why @ValstadShip exists.

Had a planning call with a major shipbuilder today. We start production this summer. This is the beginning of a complete reinvention of American shipbuilding. It won't look like much at first, but in a few years... you'll know.

Super excited to be working with the @BlueWaterShips team to advance American shipbuilding!

Research robot! Tall Bowen, our summer intern, is exploring foundation models for contact-rich assembly tasks.

One of the many awesome things happening at ProtoTown.

If you want to a) travel long distances, and b) carry a lot of stuff, you need big ships. Small ships have their place in combat, but they can’t sustain a war effort on their own.

I think often about what we’re doing in the context of the AI boom. On the one hand, AI companies are scaling at unprecedented rates - this can make the slower initial ramp of new industrials harder for VCs to invest in. On the other, funds usually have a 7-10 year lifecycle. There’s a real question about how durable some of these buzzy startups will be in 3-5 years. Like we saw in 2021, when valuations invariably correct there will be a minefield of zombie companies with a high burn rate, an ugly preference stack and limited access to new funding. New industrials get accelerated by AI, but they still need to do the hard work of moving atoms - it takes longer to ramp but once scaled they become very difficult to dislodge. Valuations tend to be more reasonable early on. In a world where startups mainline investor cash straight into model companies (like they used to with Meta ads, and still do), the capital requirements actually aren’t that crazy for deep tech / hard tech, and there are myriad non-dilutive ways to finance physical infrastructure and equipment. Venture partners can’t over-index on patriotism - their job is to deliver venture-scale returns - but they can certainly market this as an objective to LPs, many of whom will care a great deal about the strengthening of American industry. I’m sure anyone who backed SpaceX early is very happy with the returns, but almost certainly bought into the mission long before the financial outcome was certain.

Accurate.

I once pitched @pmarca while standing in line for an 1878 performance of the HMS Pinafore. He laughed maniacally and disappeared for a few moments before re-emerging in a Napoleon Bonaparte uniform, regaling the audience with a 90-minute Rachmaninoff performance on a harpsichord lowered from the ceiling. Seemingly satisfied with his deft handling of the Third Movement of Piano Concerto No. 3, he broke all the windows with his head, whispered “nobody will ever believe you” and galloped off on a magnificent stallion into the setting sun. Never meet your heroes.

Every time I have to rent a car I feel like I’ve stepped back in time 20 years. Why isn’t there a Tesla-only rental car company? And yes, I’m familiar with Turo.

Fiiiine we’ll build it.