A pair of former SpaceX engineers who once worked on Starlink and Starship now want to point the same first-principles thinking at a decidedly earthbound problem: the excavators, dozers and graders that build America's roads, substations and homes. Their startup, TerraFirma, said in mid-July 2026 that it has raised $100 million in new funding, capital that pushes its total investment to date to roughly $115 million and thrusts a two-year-old Austin company into the center of the fast-heating race to automate the construction site.
Kleiner Perkins led the round, which the company characterized as its Series A. Participating investors included Bain Capital Ventures, Glade Brook Capital Partners, BANNER VC, Saga Ventures, Trust Ventures, Definition, PEAK6, Magnetar Capital and Ravelin Capital, along with a roster of angels drawn from the ranks of SpaceX, Anduril, Base Power, Shinkei and Hadrian. The deal landed TerraFirma a spot among the week's ten largest U.S. venture rounds, according to Crunchbase, in a stretch dominated by AI-focused financings.
Robots in the cab, operators at the screen
TerraFirma, founded in 2024 by co-founders Noah Schochet and Noah McGuinness, is not selling a single robot. It is building what it calls a full-stack "robotic infrastructure" platform that pairs AI-enabled pre-construction software with a remote command-and-control center and retrofitted heavy machinery, from excavators and dozers to loaders, rollers and skid steers. The company converts conventional equipment into machines that no longer need an operator sitting in the cab. Instead, skilled operators run entire fleets from screens, extending the judgment they have built over years across several machines at once.
That distinction, autonomy as a component rather than the whole answer, is central to the company's pitch. "It is not about trying to fully automate construction equipment," McGuinness, the chief technology officer, said in the funding announcement. "Making construction truly faster and cheaper requires innovating on operations and technology together across the full stack. We believe autonomy is a part of the solution, but driving real change requires building a whole ecosystem of technology that is directly informed by rapid iteration and lessons from the field."
TerraFirma says the approach can make a single operator up to 300% more effective, a claim that, if borne out at scale, speaks directly to the industry's most stubborn problem.
Why construction is fertile ground for AI
Construction is one of the few major industries where labor productivity has gone backward. Since 1965, productivity in U.S. construction has fallen at an average rate of about 0.6% per year, even as the broader economy grew productivity at roughly 1.6% annually, according to figures from the U.S. Department of Commerce and Goldman Sachs cited by the company. TerraFirma pegs the cumulative cost of that decline at the equivalent of roughly $1 trillion every five years.
Layered on top is a chronic labor shortage. The sector has struggled for years to recruit and retain skilled equipment operators, and an aging workforce means the gap is widening. Teleoperation and semi-autonomy offer a way to stretch scarce expertise further while pulling workers out of the dust, noise and hazard of the cab, jobs the company argues can be made safer and higher-paying in the process. That combination, a real productivity crisis plus a genuine worker shortfall, is exactly why investors have grown bullish on physical AI aimed at dirt and steel rather than screens.
TerraFirma is not alone. The same Crunchbase roundup that flagged its raise also listed Walden Robotics, which launched from stealth with $300 million to build general-purpose robots for manufacturing and logistics. Rivals such as Built Robotics have pursued autonomous earthmoving for years, and smaller players like HIVE are retrofitting existing industrial machines with sensors and onboard intelligence. What sets TerraFirma apart, its backers argue, is that it operates as a vertically integrated contractor, building and using its own robotics and software on live job sites rather than selling kit to others.
The company is already booking commercial work. Recent Texas projects include site preparation, excavation and grading for a new Starbucks in North Austin, a sports arena in Spicewood and a power substation in New Braunfels. TerraFirma also says it is working with the U.S. government on infrastructure and logistics projects in challenging environments abroad.
"TerraFirma is succeeding at real-world scale, proving the business model works, and securing government and commercial contracts," said Josh Coyne, a partner at Kleiner Perkins. "This is clearly where the industry is headed."
What to watch
The near-term test is execution. TerraFirma says the new money will expand its engineering, manufacturing, operations and construction teams and fund continued development of its semi-autonomous equipment. The questions that will determine whether the round looks prescient are whether that 300% operator-efficiency figure holds outside a handful of showcase Texas sites, whether the company can hire against the same labor scarcity it is trying to solve, and whether government contracts translate into durable revenue. Schochet, for his part, is thinking well past the next grading job. "The technology that we are building today to solve critical challenges here on Earth," he said, "will be highly reusable to solve those same challenges on the Moon and Mars." For now, the proving ground is a substation outside San Antonio.
"It is not about trying to fully automate construction equipment. Driving real change requires building a whole ecosystem of technology directly informed by lessons from the field."— Noah McGuinness, Co-founder and CTO, TerraFirma