The robot hand startup founded by a former Tesla Optimus engineer resolved a high-profile IP dispute and secured fresh backing from First Round Capital, signaling that investors aren't scared off by legal clouds hanging over contested spinouts.
Zhongjie "Jay" Li left Tesla in September 2024 after two years working on the advanced robotic hand sensors at the heart of the Optimus humanoid robot program. Less than a year later, Tesla filed suit against him, alleging he had downloaded sensitive design files, source code, specifications, and prototype videos to personal devices before walking out the door. The complaint painted a familiar picture in Silicon Valley: a key engineer goes, proprietary knowledge goes with him, and a competitor springs up in the gap.
This week, as TechCrunch exclusively reported, that lawsuit was dismissed following a settlement between Li and Tesla. On the same day, Proception announced it had raised an $11 million seed round led by First Round Capital partner Bill Trenchard, with Y Combinator and early-stage fund BoxGroup also participating. The timing wasn't accidental. You don't go out to raise $11 million while a trade secret case is hanging over your head if you can avoid it. The settlement cleared the runway.
Trenchard put it plainly: Proception will have "the best hand in the market, maybe the most sophisticated hand today, and the underlying data and models to support that." That's a confident claim to make about a company that, until recently, was defending itself in court against the company whose technology it allegedly copied. But investors in this space are clearly willing to hold their nose on the provenance question if the underlying product is compelling enough.
The core product is a sensor-laden glove with 22 degrees of freedom and multiple joints per finger, designed to capture human hand interaction data without requiring a robot in the loop at all. That's a meaningful technical distinction. Instead of training robot hands by strapping them to a humanoid frame and running expensive trials, Proception's approach lets researchers and robotics companies collect dexterous manipulation data from human hands directly. The company is now shipping its first batch of high-dexterity robotic hands to researchers and opening up to wider orders.
The goal, according to Li, is to become the default hand supplier to companies that don't want to spend the time or money developing their own dexterous manipulation stack. That market is real. The humanoid robot sector is moving fast enough that most companies would rather buy a solved hand than build one, and Proception is betting it can fill that gap before the incumbents catch up.
The IP battleground nobody's talking about honestly
Here's the thing nobody is saying directly: the Proception case is one episode in what is becoming a structural tension in humanoid robotics. Tesla, which has poured enormous resources into Optimus, is converting its Fremont factory for Optimus production with mass manufacturing targeted to begin between July and August 2026. Protecting the engineering knowledge embedded in that program matters enormously to the company's competitive position.
At the same time, the engineers who built those systems are the most sought-after people in the industry. Cross-disciplinary talent capable of integrating robot hardware, AI models, and production line systems is genuinely scarce, and every well-funded humanoid robotics startup is chasing the same pool. When that talent walks out of a place like Tesla and raises money the next week, you get exactly the dynamic that played out here: a lawsuit, months of legal pressure, a settlement reached on undisclosed terms, and then a fundraise timed to follow it immediately.
The settlement terms are confidential, which means nobody outside the room knows what Li agreed to, what he gave up, or whether Proception's product roadmap had to change as a result. Investors putting $11 million into a company under those conditions are implicitly accepting that uncertainty. First Round's Trenchard presumably ran diligence on what the settlement required, but the rest of the market doesn't have that visibility. That's a due diligence gap that will become more common as more Optimus engineers leave and more spinouts emerge.
China's dominance in the humanoid sector, with nearly 90 percent of all humanoid robots sold globally in 2025 coming from Chinese manufacturers, adds a layer of strategic urgency that makes American companies more aggressive, not less, about protecting what they have. Tesla isn't going to stop filing suits. And engineers who have spent years inside programs like Optimus aren't going to stop founding companies. The Proception case probably won't be the last time those two forces collide in court before settling on an undisclosed Friday afternoon.
For founders thinking about leaving a major robotics program to start something in the same space, the lesson from Proception isn't that it can't be done. It's that the legal bill and the distraction are real, the settlement terms will constrain you in ways you can't disclose, and you'd better have a product compelling enough that First Round will still write you a check on the other side of it.
Also read: OpenAI scrambles to fix Codex as coding agent usage blows past its own capacity models • Asian AI rivals fill the gap left by Anthropic's export ban and match frontier performance for a fraction of the cost • Millennium Management is building its own AI lab and that changes the competitive calculus for quant finance
