The Founders Who Left Tesla to Rebuild America | a16z
Erin Price-Wright, Turner Caldwell (Mariana Minerals), and Drew Baglino (Heron Power) discuss closing America's critical minerals gap and modernizing the power grid for the AI economy. They cover how automation, reinforcement learning, and lessons from Tesla can accelerate mining, refining, and grid infrastructure development to compete with China and enable re-industrialization.
The advancement of AI and the re-industrialization of America are fundamentally physical projects, constrained not by algorithms but by atoms. Dominance in AI requires a robust foundation of materials, energy, and a modernized electrical grid capable of moving power where and when it’s needed. The current challenge is a call to action to rebuild the nation’s industrial backbone, from mineral extraction to energy transmission.
The Bottleneck: A 50-Year Deficit in Minerals and Grid Tech
The United States finds itself significantly behind its geopolitical rivals, particularly China. This deficit is twofold:
- Critical Minerals: The US is approximately 50 years behind China in the critical mineral supply chain. The core issue isn’t just about permitting new mines; it’s that the nation is too slow in the subsequent phases. Turner Caldwell points out, “…we are too slow at designing, building, and ramping up new minerals capacity even after we have licensed to operate.” A new project can take five years to build and another three to five years to reach full operational capacity.
- Grid Infrastructure: The electrical grid is built on mechanical systems developed over a century ago. Drew Baglino notes that while innovation has surged at the grid’s edge (EVs, energy storage), the core infrastructure of transformers and transmission has seen “really been no change.” This legacy system is fragile, overbuilt, lacks modern monitoring and control, and relies heavily on equipment from overseas suppliers, posing a security risk.
Modernizing Minerals with a Software-First Approach
To close the mineral gap, Mariana Minerals is pioneering a “software-first” model for mining and refining. This isn’t a SaaS company; they are vertically integrated operators who build and run their own projects, including a copper mine in Utah and a forthcoming lithium refinery in Texas. Their strategy is to use software and autonomy to radically accelerate project delivery and operational efficiency.
Their technology stack includes three core operating systems:
- Capital Project OS: A product lifecycle management tool that uses agentic workflow automation to speed up the entire process from mine development through engineering and construction.
- Plant OS: This system uses reinforcement learning (RL) to automate refinery control. By removing humans from the loop, it can continuously tune variables like temperature, flow rates, and chemical additions to handle heterogeneous feedstock—a crucial advantage when the domestic labor pool lacks the deep, embedded know-how to operate complex refining circuits.
- Mine OS: Also leveraging RL, this OS enables short-interval autonomous control of mining operations, optimizing the thousands of daily decisions to maximize productivity and equipment utilization.
By embedding software engineers with operating teams, Mariana aims to solve the core problems on the ground and drive technology adoption at a rate traditional companies cannot match.
Upgrading the Grid with Silicon and Software
Heron Power is focused on bringing the grid into the modern era by replacing “steel, oil, and copper” with “silicon and software.” The company builds solid-state transformers, leveraging decades of advancements in power transistors that have paralleled Moore’s Law.
This technology offers a path to a more resilient, controllable, and efficient grid. Baglino argues that since the US pioneered and leads in producing key power semiconductors like silicon carbide, it should be the first to commercialize and manufacture the applications at home. Failing to do so means ceding the immense benefits of this foundational technology to other countries.
The Tesla Playbook for Industrial Innovation
Both founders bring critical lessons from their time at Tesla, a company that provided a template for rebuilding American manufacturing. The “Tesla model” provides several key advantages over traditional industrial companies:
- Techno-Optimism: A core belief that old, archaic systems can be fundamentally re-imagined and innovated upon.
- High Appetite for Risk: This cultural trait enables extremely fast decision-making and empowers teams to move quickly without fear of failure.
- Relentless Persistence: As Baglino puts it, “If the outcome is worth it, Tesla will fight through the challenges of getting to that outcome.” This contrasts with companies that shelve innovation projects after initial setbacks.
- A Mission-Driven Talent Magnet: Tesla’s clear and ambitious purpose attracts the best talent. The high-growth environment provides rapid career advancement, which helps retain them. This creates a focused, “do or die” culture that is difficult to replicate in a legacy industrial conglomerate.
Rebuilding the American Workforce and Supply Chain
A common argument against onshoring is labor cost, but in modern, automated factories, the labor differential between the US and China can be less than 5% of the cost of goods sold. The real competitive driver is the supply chain. China excels at creating dense, co-located industrial zones where thousands of components are available within a few hours’ drive, drastically reducing logistics time and cost. Replicating this model is a major unlock for the US.
To build the necessary workforce, founders must be creative and look to analogous industries. For battery manufacturing, Tesla hired from high-speed bottling plants; for mining, Mariana recruits from oil and gas. The software talent for complex optimization algorithms can be found in fields like ad-tech and logistics, demonstrating the transferability of skills from the digital economy to the physical one.
A Call for Durable Industrial Policy
To accelerate this industrial revival, the founders made several specific policy asks:
- A Minerals Mandate: Create a long-term strategic framework for critical minerals, similar to the “energy mandate” that supported oil and gas for 50 years. This requires creating the right incentive structures to assure private capital markets of long-term stability.
- Durable Industrial Policy: Enact predictable, long-term policies that companies and their suppliers can plan around.
- Coordinated Infrastructure Planning: A concerted effort between federal and state governments to identify dedicated zones for energy and manufacturing buildouts. This would help create co-located supply chains and align local jurisdictions to say “yes” to critical projects.
- A Federal Grid Investment Framework: Baglino proposed a “federal highway trust fund for the grid”—a master plan to finance and build out linear electrical infrastructure, connecting new industrial zones and improving national resilience.