Figure AI — humanoid robots for the commercial workforce
Figure AI is a Sunnyvale-founded, San Jose-headquartered robotics company building general-purpose humanoid robots for commercial deployment. With Figure 03 — the company's 2025 flagship designed for mass manufacturing — Figure is positioning itself among the leading players of the current humanoid robotics and AI boom, alongside Tesla Optimus, 1X, Apptronik, and Boston Dynamics.
- 2022Founded
- San Jose, CAHeadquarters
- Figure 03Latest flagship (2025)
- $1B+ Series CFunding raised
- USACountry
- PrivateStatus
About Figure AI
Figure AI, Inc. was founded in 2022 by serial entrepreneur Brett Adcock — previously the co-founder of Vettery and Archer Aviation — with the singular goal of building a general-purpose humanoid robot for the commercial workforce. Headquartered initially in Sunnyvale, California and later relocating its main operations to San Jose, the company set out to solve the two hardest problems in humanoid robotics simultaneously: the mechatronics of a bipedal, dexterous machine, and the general-purpose AI required to make it useful in real, unstructured work environments.
The company's public trajectory has been aggressive. Its first prototype — Figure 01 — walked untethered in 2023, roughly 18 months after founding. In early 2024, a widely shared demonstration featured Figure 01 conversing and performing kitchen tasks, powered in part by an OpenAI language-and-vision partnership. Later that same year, Figure began piloting its Figure 02 platform on the BMW Group's assembly line in Spartanburg, South Carolina — the first announced commercial deployment of a humanoid robot by a major global automaker.
In 2024 Figure also unveiled Helix, its proprietary Vision-Language-Action (VLA) model, and formally ended its OpenAI collaboration in favor of in-house AI. The move reflected a broader thesis: that owning both the hardware and the foundation model is the durable moat in humanoid robotics. Backed by a Series B round that valued the company at approximately $2.6 billion and a subsequent Series C reported to exceed $1 billion — with investors including Microsoft, NVIDIA, OpenAI, Jeff Bezos, Intel Capital, and others — Figure is reportedly valued at approximately $39.5 billion following its most recent round, making it one of the most highly capitalized humanoid robotics startups in the world.
The team, currently more than 200 employees and weighted heavily toward hardware, controls, and AI engineers, has grown rapidly since founding. Adcock has publicly stated that Figure's long-term ambition is not merely to build a factory robot but to build a general-purpose humanoid capable of handling the majority of physically demanding, high-turnover jobs across the global economy — a vision consistent with the broader current AI boom in general-purpose robotics, which now includes Tesla Optimus, 1X NEO, Apptronik Apollo, Agility Digit, and Boston Dynamics Atlas as direct competitors.
Product lineup: Figure 01, Figure 02, and Figure 03
Figure has iterated on a roughly annual cadence since its first hardware prototype. Each generation has meaningfully narrowed the gap between research demonstration and factory-ready product, culminating in the 2025 Figure 03 — the company's first platform explicitly designed for mass manufacture at a target rate of more than 100,000 units per year.
Figure 01 (2023) — the research prototype
Figure 01 was the company's first ambulatory humanoid: a research-grade platform used to validate the basic mechatronics, walking gait, and manipulation stack. It became widely recognized for its OpenAI-powered speech-and-vision demonstration in early 2024.
Figure 02 (2024) — the commercial pilot
Figure 02 was a substantial redesign focused on production-line readiness: improved actuators, onboard compute, higher payload, longer runtime, and a re-worked exterior. It is the platform piloted at BMW Group Plant Spartanburg, where it is being evaluated for assembly tasks.
Figure 03 (2025) — the mass-manufacturable flagship
Figure 03 is the company's most significant product step so far. It runs on the in-house Helix VLA model, drops external AI dependencies, and — most importantly — is engineered around a supply chain and BOM designed for high-volume production. Figure has publicly targeted an annual manufacturing capacity exceeding 100,000 units, an order of magnitude larger than any humanoid robot ever produced at scale.
| Model | Year | Height | Weight | Payload | Battery | Actuators | Notable |
|---|---|---|---|---|---|---|---|
| Figure 01 | 2023 | ~1.68 m | ~60 kg | ~20 kg | ~5 hr | Electric BLDC | OpenAI-powered demo; R&D prototype |
| Figure 02 | 2024 | ~1.68 m | ~70 kg | ~20 kg | ~5 hr | Redesigned BLDC | BMW Spartanburg pilot; commercial platform |
| Figure 03 | 2025 | ~1.68 m | ~60 kg | ~20 kg | ~5 hr | Next-gen BLDC | Helix VLA in-house; mass-manufacturable, >100K/yr target |
Technology stack
Figure's platform is a vertically integrated combination of custom mechatronics, sensor fusion, high-performance onboard compute, and a proprietary foundation model. The company has publicly framed the humanoid problem as one that requires both best-in-class hardware and a dedicated general-purpose AI model — and it now builds both.
Perception
Figure robots use a multi-camera vision system for full 360-degree situational awareness, supplemented by force-torque sensing and tactile feedback at the manipulators. This mix of vision and touch is what allows the robot to perform contact-rich tasks — inserting parts, gripping deformable objects, and reacting to unexpected contact — in an unstructured factory or warehouse environment.
Helix VLA model
Helix is Figure's in-house Vision-Language-Action model. Publicly disclosed as a roughly 7-billion-parameter system running whole-body control at up to 200 Hz, Helix maps natural-language instructions and camera input directly to actuator commands. Because the model runs on the robot itself, Helix does not require cloud connectivity for real-time tasks — an important requirement for factory-floor deployment.
Actuators, power, and compute
Figure uses electric brushless DC (BLDC) actuators throughout the body — no hydraulics — which simplifies serviceability and improves efficiency. Each robot carries a custom battery pack targeting approximately five hours of continuous operation, and the onboard compute stack is understood to include NVIDIA-class GPU acceleration for the Helix model. The result is a self-contained platform: perception, planning, control, and power all live on the robot.
Power management is a particularly demanding subsystem for a humanoid of this class. Each joint's servo controller must deliver hundreds of watts of instantaneous torque without thermally throttling, which drives the use of high-efficiency wide-bandgap semiconductors (SiC and GaN) in the motor-driver stages and requires tightly matched precision passives on every gate-drive and current-sense channel. The onboard battery pack integrates cell balancing, thermal management, and multi-stage protection ICs — subsystems whose reliability is often the practical ceiling on how long a humanoid platform can run between charges.
Applications and commercial deployments
Figure's stated market is the commercial workforce — jobs in manufacturing, logistics, and warehousing that are physically demanding, high-turnover, or otherwise difficult to staff. The company's first announced customer is a marquee one:
- BMW Group — Plant Spartanburg (South Carolina). Figure 02 robots have been evaluated on the automotive assembly line since 2024, handling sheet-metal parts and related manual tasks. This makes BMW the first major global automaker to publicly test humanoid robots in a production environment.
- Logistics and warehousing (GXO — reported). Multiple industry reports have named GXO Logistics as an additional pilot customer, with Figure targeting case-handling, pick-and-place, and general material-movement tasks common in modern distribution centers.
- Additional commercial pilots. Figure has signaled interest in other heavy-industry and high-turnover sectors, and the Figure 03 platform is explicitly positioned to make those broader rollouts economically feasible.
The robotics supply chain — where Hybrid Electronics fits in
A modern humanoid robot is one of the most component-dense products ever built. A single Figure-class platform depends on thousands of individual electronic components: precision BLDC motor drivers, high-torque servo modules, IMUs and 6-DoF force-torque sensors, LiDAR and time-of-flight modules, GPU compute boards, battery-management ICs, wide-bandgap SiC/GaN power semiconductors, high-speed board-to-board connectors, precision resistors, and thousands of MLCC capacitors distributed across every power and signal rail.
Building that kind of hardware puts enormous pressure on the electronic-components supply chain. Prototype pilots and low-volume production runs frequently hit allocation, EOL notices, and long lead times on the exact parts they need, and R&D and field-service teams routinely have to support older platforms that use semiconductors no longer in mainstream distribution. During the current humanoid robotics and general-purpose robot AI boom, this pressure is amplified: dozens of well-funded teams are chasing the same short list of high-torque BLDC drivers, precision IMUs, GPU compute modules, and automotive-grade power semiconductors, and franchise stock on many key parts is now measured in months of lead time rather than weeks.
Hybrid Electronics has been sourcing electronic components for industrial and embedded systems since 1995. For robotics teams — humanoid, mobile, industrial, and research — we specialize in finding the parts you need when mainstream distribution can't:
Obsolete and end-of-life parts
Legacy semiconductors, sensors, connectors, and passives that are no longer in mainstream distribution — the parts that keep older platforms running during transitions to newer hardware, and the long-tail items that block a BOM from closing.
Allocated and hard-to-find active parts
High-torque BLDC motor drivers, IMUs and MEMS sensors, SiC/GaN power devices, GPU compute modules, precision resistors, and MLCC capacitors caught in allocation or long-lead times. Send us a BOM or a shortage list — we'll confirm what we can source and at what lead time.
Frequently asked questions
Who founded Figure AI?
Figure AI was founded in 2022 by Brett Adcock, previously the co-founder of the recruiting marketplace Vettery and the electric-aircraft company Archer Aviation. Adcock remains CEO. The company is headquartered in San Jose, California.
What is Figure 03?
Figure 03 is the 2025 flagship general-purpose humanoid robot from Figure AI. It is the company's first platform explicitly engineered for mass production, targeting more than 100,000 units per year, and it runs on the in-house Helix Vision-Language-Action model rather than any third-party foundation model.
Where are Figure robots deployed commercially?
The most prominent announced deployment is at BMW Group Plant Spartanburg in South Carolina, where Figure 02 has been evaluated on the automotive assembly line since 2024. Industry reports have also named GXO Logistics as an additional pilot customer for warehousing and material-movement tasks.
How much has Figure AI raised?
Figure closed a Series B in February 2024 that valued the company at approximately $2.6 billion post-money, with investors including Microsoft, NVIDIA, OpenAI, Jeff Bezos, and Intel Capital. It has since raised a Series C reported to exceed $1 billion, with the company reportedly valued at approximately $39.5 billion following that round.
Is Hybrid Electronics affiliated with Figure AI?
No. Hybrid Electronics is an independent electronic-components sourcing company. We are not a distributor, reseller, or partner of Figure AI. All trademarks are property of their respective owners. Hybrid supports robotics developers by sourcing obsolete, allocated, and hard-to-find semiconductors, sensors, connectors, and passives for R&D, prototype, and low-volume production builds.
Can Hybrid help with components for humanoid robotics builds?
Yes. We regularly support R&D and integrator teams with hard-to-find semiconductors, BLDC motor-driver ICs, IMUs and other MEMS sensors, connectors, SiC/GaN power devices, and precision passives for the industrial and embedded boards used in robotics platforms. Send us a BOM or a short list of shortages and we'll confirm availability and pricing.
Trademark and affiliation notice. Figure AI®, Figure 01, Figure 02, and Figure 03 are trademarks of Figure AI, Inc. Hybrid Electronics is an independent electronic-components sourcing company and is not affiliated with, endorsed by, or authorized by Figure AI, Inc. All information on this page is compiled from public sources for educational and reference purposes.