Steinmetz Is Making Motor Controllers That Embarrass the Competition on Power Density

The Macro: The EV Supply Chain Needs Better Components

The electric vehicle industry has a parts problem. Everyone talks about batteries. Batteries get the headlines, the billion-dollar factory announcements, the geopolitical drama about lithium supply chains. But the motor controller, the component that converts DC battery power into the AC signals that actually spin the motor, has been quietly stuck in the same performance range for years.

Motor controllers (also called inverters) are one of those components that most people outside the industry never think about. They sit between the battery and the motor. Their job is to deliver power efficiently, handle thermal loads, and do it all in a package small enough to fit in increasingly tight vehicle architectures. The dominant players are BorgWarner, ZF, Vitesco, and the in-house teams at major automakers. Innovation has been incremental. Efficiency gains of half a percentage point get celebrated.

This matters because power density, how much power you can push through a given volume of hardware, directly affects vehicle design. A smaller inverter means more room for passengers, or a smaller overall vehicle footprint, or less cooling infrastructure. In robotics and aviation, where weight and space constraints are even tighter, power density is the entire ballgame.

The silicon carbide transition has helped. SiC switching devices run at higher frequencies and temperatures than traditional silicon IGBTs, which enables smaller, more efficient inverter designs. But most existing manufacturers are retrofitting SiC into old architectures. The real gains come from designing the whole system around SiC from scratch.

The Micro: Tesla and SpaceX Alumni Building Inverters

Steinmetz was founded by Owen Brake and Ethan Childerhose. Owen built motor controllers at Tesla. Ethan worked at Rocketlab, Nvidia, Tesla, and Neuralink. They are a two-person team in San Francisco, part of Y Combinator’s Winter 2025 batch. The company is named after Charles Steinmetz, the electrical engineering pioneer who basically invented AC motor theory in the early 1900s.

Their flagship product is the MC0-500, a 500kW motor controller that claims 104 kilowatts per liter of power density. For context, the Tesla Model 3 inverter runs at about 65 kW/L. If Steinmetz’s numbers are real, that is a 60% improvement, which is massive in a field where 10% gains take years of R&D.

They have three products in their 0-Series line. The MC0-500 is shipping now. The MC0-250 (250kW, aimed at industrial applications) ships in March 2026. The MC0-10 (15kW, targeting robotics and automation) is coming in Q3 2026. All claim greater than 99% efficiency across the line.

The product details are interesting. Open-source firmware is unusual for power electronics hardware. One-click auto-tuning means customers do not need a power electronics PhD to commission the thing. Multi-sensor support and real-time monitoring suggest they are building for teams that want to integrate quickly rather than spend months on custom configurations.

Manufacturing is in America, which matters for EV companies dealing with IRA domestic content requirements and for defense and aerospace customers who cannot source from overseas.

The pricing is not public. For hardware at this performance level, I would expect it to sit above commodity inverters but well below custom aerospace solutions. The sweet spot would be customers who need high performance but cannot justify a full custom engineering program.

The Verdict

I think Steinmetz is going after a genuinely underserved market. The major inverter suppliers are focused on high-volume automotive programs. If you are building a specialty EV, a robotics platform, or an aerospace application, your options are limited. You either use an off-the-shelf unit that does not quite fit or you spend a fortune on custom development.

The power density claim is the whole pitch. 104 kW/L versus 65 kW/L is the kind of number that makes engineers stop scrolling. If it holds up under real-world thermal and duty cycle conditions, Steinmetz has a product that sells itself to the right buyers.

The risk is hardware. Hardware startups are hard in ways that software startups are not. You need supply chain management, manufacturing partnerships, quality control, and the ability to handle warranty claims. A two-person team shipping power electronics is impressive and also a little terrifying. Scaling from prototype-grade production to consistent volume output is where hardware companies go to die.

Competitors in the startup space include Cascadia Motion (which BorgWarner acquired) and various SiC inverter programs at the major Tier 1 suppliers. On the small-batch side, companies like Rinehart Motion Systems serve the specialty EV market. None of them are claiming 104 kW/L.

Thirty days, I want to see how many MC0-500 units have shipped and to whom. Sixty days, the question is whether the MC0-250 hits its March ship date. Ninety days, I want to hear from customers running these in real vehicles or robots. Do the power density numbers hold up outside the lab? Hardware claims are only as good as the field data behind them.