The hardest part of building a useful robot isn't the intelligence. It's the hands. For all the progress in large language models and vision systems, getting a machine to pick up a screwdriver, thread a needle, or assemble a small electronic component remains a stubbornly physical challenge. Kyber Labs, a Brooklyn-based robotics startup, is betting its novel artificial muscle fiber actuators are the missing piece. The company is building a modular hand-based manipulation platform designed from the ground up to be controlled by AI, aiming to bring human-like dexterity to tasks that have resisted automation.
A hardware wedge for embodied AI
Kyber's core proposition is an actuator technology that mimics the properties of human muscle. Unlike the rigid, high-torque motors common in industrial robots, these artificial muscles are designed to be compliant, low-cost, and capable of fluid, nuanced movement [Kyber Labs site, retrieved 2024]. The company's platform focuses on the hand, a modular unit that can be mounted on existing robotic arms. This approach lets them sidestep the immense complexity and cost of building a full humanoid robot, instead targeting the specific problem of dexterous manipulation in unstructured environments [Kyber Labs site, retrieved 2024]. The wedge is clear: provide the physical interface that allows increasingly sophisticated AI control models to interact with the real world.
The team and the early traction
The founding team brings a mix of deep tech and hardware experience. Co-founder and COO Yonatan Robbins is a SpaceX veteran, having worked on flight reusability systems and mechanisms for the Falcon 9, Falcon Heavy, and Starship programs [Kyber Labs site, retrieved 2024]. Co-founder and CEO Tyler Habowski has a background spanning Endless Frontier Labs and Machina Labs [RocketReach, retrieved 2026]. They've assembled a small team that includes robotics researcher Julian Meier-Viereck [pr.ai thread]. This technical pedigree helped secure a $1.7 million pre-seed round in late 2023, with backing from a consortium of deep tech and frontier technology investors including Cortical Ventures, Endless Frontier Labs, and Y Combinator [Deep Tech Week, September 2023].
| Role | Name | Notable Background |
|---|---|---|
| Co-Founder, CEO | Tyler Habowski | Endless Frontier Labs, Machina Labs, SpaceX [RocketReach, retrieved 2026] |
| Co-Founder, COO | Yonatan Robbins | SpaceX (F9/FH/Starship mechanisms) [Kyber Labs site, retrieved 2024] |
| Founding Robotics Researcher | Julian Meier-Viereck | Kyber Labs [pr.ai thread] |
The competitive landscape for dexterity
Kyber Labs is not alone in chasing robotic manipulation. Competitors like Clone Robotics and MetisMotion are also developing advanced robotic hands and actuation systems. The broader field is crowded with well-funded humanoid robot companies, though Kyber's focus on a modular hand platform sets it apart from those pursuing full-body systems. The company's technical differentiator rests on the performance and cost profile of its artificial muscle fibers, which it claims enable movement not feasible with conventional systems [Automate.org, retrieved 2026]. The bet is that this specific hardware innovation creates a defensible moat, even as AI software for control rapidly advances.
What could go wrong at scale
The path from a promising actuator in a lab to a reliable, cost-effective product deployed in a real warehouse or factory is long and fraught. Hardware development cycles are measured in years, not months, and scaling manufacturing for a novel material presents its own set of challenges. Kyber's technology must prove it can withstand millions of cycles, operate consistently across temperature ranges, and be produced at a price point that makes economic sense for customers. Furthermore, the value of a superior hand is only realized if the AI control stack is sophisticated enough to exploit it. The company is betting on the parallel maturation of both hardware and software, a coordination problem that has tripped up many robotics ventures.
From an engineering standpoint, the technical hurdles are significant. The actuator's force density, hysteresis, and power efficiency under load will determine its practical utility. Long-term durability testing in abrasive, real-world environments is a non-negotiable step that comes later in the development cycle. The company's current $1.7 million war chest [Deep Tech Week, September 2023] will need to be significantly expanded to fund that journey from prototype to production.
The market they are targeting,automation for complex assembly and manipulation to address labor shortages [Kyber Labs LinkedIn, retrieved 2024],is vast, but also one where incumbents with established supply chains and sales relationships dominate. Kyber's success hinges on proving its platform is not just technically superior, but also simpler and more cost-effective to integrate than retooling an entire production line. For now, the company is hiring for critical roles like Staff Engineer and Head of Engineering [Y Combinator, retrieved 2026], a signal that the focus is on building the core technology before tackling commercialization at scale.
Sources
- [Kyber Labs site, retrieved 2024] Company description and technology | https://kyberlabs.ai/
- [Kyber Labs LinkedIn, retrieved 2024] Mission statement | https://www.linkedin.com/company/kyber-labs
- [Deep Tech Week, September 2023] Pre-seed funding announcement | https://www.deep-tech-week.com/organizations/kyber-labs
- [RocketReach, retrieved 2026] Tyler Habowski background | https://pitchbook.com/profiles/company/534548-80
- [pr.ai thread] Team information | https://pr.ai/threads/kyber-labs-inc-robotics-engineering-brooklyn-new-york-new-york-usa.27744/
- [Automate.org, retrieved 2026] Company profile and technology claims | https://www.automate.org/companies/kyber-labs
- [Y Combinator, retrieved 2026] Open job listings | https://www.ycombinator.com/companies/kyber/jobs/GPJkv5v-staff-engineer-tech-lead