⚠ Disclaimer: This entry may be incomplete, out of date, or inaccurate. It is AI-maintained on a best-effort basis. Do not rely on it as a sole source — verify claims independently using the sources listed below.
Summary
Apptronik is an Austin-based humanoid robotics company spun out of the Human Centered Robotics Lab at UT Austin, with deep roots in NASA’s Valkyrie humanoid program. Its Apollo robot targets industrial manufacturing and logistics. The company raised $935M total across a multi-tranche Series A in 2025–2026, co-led by B Capital and Capital Factory with Google participation, at a ~$5B post-money valuation. Investors include Mercedes-Benz, John Deere, Qatar Investment Authority, and AT&T Ventures. Apollo features a 4-hour battery life with hot-swap capability (swap in under 5 minutes), directly addressing the runtime limitation that constrains competitors. Commercial partners with active Apollo deployments include Mercedes-Benz, GXO Logistics, and Jabil.
Key Facts
- Founded: 2016 (UT Austin Human Centered Robotics Lab spinout)
- HQ: Austin, TX
- Type: Company — Platform OEM
- Status: Active; late pre-commercial / early commercial
- Employees: ~300 (as of February 2026)
- Total funding: ~$935M
- Valuation: ~$5.3B post-money (February 2026 round)
- Robot: Apollo — 5'8" (~1.73m), ~160 lbs (~73 kg), 25 degrees of freedom, 55 lb (25 kg) payload
- Battery: 4-hour runtime; hot-swappable in under 5 minutes
- Actuator technology: UT Series Elastic Actuator (SEA), invented by Nicholas Paine
- Value chain position: Platform OEM; actuator design in-house (SEA), components sourced
- Commercial partners (active deployments): Mercedes-Benz, GXO Logistics, Jabil
- Key investors: B Capital (lead), Capital Factory (co-lead), Google, Mercedes-Benz, John Deere, Qatar Investment Authority, AT&T Ventures, PEAK6
Funding & Investors
| Round | Date | Amount | Lead / Notable Investors |
|---|---|---|---|
| Pre-Series A (NASA grants) | 2016–2024 | $5M+ | NASA / government contracts |
| Series A initial | Feb 2025 | $350M | B Capital (co-lead), Capital Factory (co-lead), Google (participant), Mercedes-Benz, GXO Logistics |
| Series A extension | Mar 2025 | $53M | Oversubscribed close; brings initial round to $403M |
| Series A-X extension | Feb 2026 | $520M | Existing investors including B Capital, Google, Mercedes-Benz; new: AT&T Ventures, John Deere, Qatar Investment Authority (QIA) |
Total raised: ~$935M+ across the full Series A. The Feb 2025 press release confirmed 150 employees at close; ~300 by Feb 2026.
Lead investors: B Capital (Howard Morgan, Chair; co-led initial) and Capital Factory (Gordon Daugherty, Chairman of Apptronik board) — not Google. Google participated in the initial tranche and joined the extension, but was not the lead.
Google DeepMind connection: Google’s participation across multiple tranches is widely interpreted as DeepMind seeking to ground its physical AI research in a hardware platform. Apptronik established a formal strategic partnership with Google DeepMind before the funding round. This makes Apptronik a potential hardware partner in Google’s foundation-model-for-robotics strategy.
Qatar Investment Authority: QIA is a sovereign wealth fund with $475B+ in AUM. Their participation signals interest in diversifying into frontier technology through a US ally’s manufacturing platform — and potentially Middle Eastern deployment of industrial robots.
Mercedes-Benz: Both investor and potential customer. Mercedes has publicly stated interest in testing humanoids for small-batch and precision assembly tasks at automotive plants — the same manufacturing context as GXO/Toyota are for Agility.
John Deere: Agricultural equipment manufacturer. Their investment is the most unusual — suggests interest in humanoids for farm equipment assembly or eventually field operations, although the latter is speculative.
What It Is / How It Works
Apptronik evolved from UT Austin’s Human Centered Robotics Lab, which contributed significantly to NASA’s Valkyrie humanoid (developed 2013–2015 for DARPA’s Robotics Challenge). The Apollo robot inherits Valkyrie’s emphasis on safe human co-location, modular design, and the Series Elastic Actuator technology invented by Nicholas Paine during his PhD work.
UT Series Elastic Actuators: Nicholas Paine’s doctoral dissertation produced a class of SEA designs specifically optimized for full-body humanoid robots. The UT SEA is designed around backdrivability (the robot can be pushed away in an emergency), torque sensing via spring deflection, and shock tolerance. Like Agility’s Digit, Apollo prioritizes safe human interaction over maximum speed or stiffness — the right tradeoff for factory floor co-working.
Apollo specifications: 5'8" tall (~1.73m), ~160 lbs (~73 kg), 25 degrees of freedom, 55 lb (25 kg) payload capacity, 4-hour battery runtime with hot-swappable batteries (swap in under 5 minutes). The 4-hour runtime is a significant differentiator over Agility Digit’s 90-minute limit; hot-swap means a single charging station can sustain near-continuous operation across shifts without the robot going offline. The 55 lb payload is meaningfully higher than Digit’s tote-transport capacity, positioning Apollo for heavier manufacturing tasks (automotive parts, equipment assembly).
AI integration with Google: The Google investment is likely coupled with access to Google DeepMind’s Gemini Robotics foundation models. Apptronik would use these models to accelerate task learning across manufacturing environments, reducing the teleoperation data collection burden that limits competitors. This is a significant potential competitive advantage if DeepMind’s physical AI models mature.
NASA Valkyrie lineage: Apptronik’s engineers built, operated, and maintain NASA Valkyrie units at multiple research institutions. This gives the team more full-humanoid integration experience than any other humanoid company except Boston Dynamics — the experience of actually running a full humanoid at scale in research environments is not trivial.
Founder Backgrounds
Jeff Cardenas — Co-founder & CEO
- M.S., Technology Commercialization, UT Austin
- Background in management consulting (Deloitte) before joining the robotics lab
- Bridges research-to-commercial translation; led Apptronik’s pivot from contract R&D work to product company
- LinkedIn: jeffcardenas
Nicholas Paine — Co-founder & CTO
- Ph.D., Electrical and Computer Engineering, UT Austin
- Invented the UT Series Elastic Actuator during his dissertation — the core actuator technology in Apollo
- Previously: research engineer at NASA Johnson Space Center contributing to Valkyrie hardware
- One of the small number of people globally who have designed, built, and operated a full-scale humanoid
- LinkedIn: nickpaine
Bill Helmsing — Co-founder
- Ph.D. or advanced degree in mechanical/robotics, UT Austin affiliation
- Mechanical design lead
Bill Welch — Co-founder
- UT Austin Human Centered Robotics Lab
- Controls and systems engineering
Prof. Luis Sentis — Co-founder
- Director, Human Centered Robotics Lab, UT Austin (founding academic)
- Ph.D. from Stanford; decades of whole-body humanoid control research
- Author of foundational papers on whole-body motion control that underpin modern humanoid designs
- LinkedIn: luis-sentis
Pattern: The strongest academic foundation of any US humanoid startup. Five co-founders with PhDs and genuine full-humanoid construction experience. The presence of Prof. Sentis as a co-founder (rather than just an academic advisor) gives Apptronik access to ongoing research output and talent pipeline from UT Austin.
People — Last Reviewed: 2026-06-19
Supply Chain & Dependencies
Actuators (40–60% of BoM): UT SEA design; components (motors, springs, bearings, gearboxes) sourced externally. Motors likely from Maxon (Swiss, premium) or Kollmorgen; precision bearings potentially from Schaeffler or Japanese suppliers. Unlike Figure AI, Apptronik does not claim full actuator vertical integration.
Compute: Not publicly specified. Given Google investor/partner status, Tensor Processing Units (TPUs) for cloud training and NVIDIA Jetson-class for edge inference are the likely stack. Google may provide compute infrastructure as part of the partnership.
Sensors: Camera + depth sensor head; lidar not mentioned. Likely RGB-D sensors from Luxonis or equivalent; proprioceptive sensing via SEA spring deflection.
Battery: 4-hour runtime; hot-swappable packs replaceable in under 5 minutes. This is Apollo’s most important operational specification — it enables near-continuous shift operation with a small bank of spare batteries, unlike competitors requiring full recharging downtime. Cell chemistry and supplier not disclosed.
Manufacturing: Not discussed publicly. No BotQ-equivalent announced. Production scale is small; commercial scale-up would require a manufacturing partner (Mercedes-Benz or another strategic investor could provide this).
Supply chain geography: No explicit China decoupling statement. Given US manufacturing ambitions and the John Deere/Mercedes-Benz investor base, domestic supply chain is likely a priority but not yet documented.
Claim Verification
Claim: Apollo will reach commercial manufacturing deployment at scale
Status: Unverified (forward-looking)
Supporting sources:
- $935M in funding with Mercedes-Benz and John Deere as strategic investors signals near-term deployment intent
- Google DeepMind partnership suggests access to foundation AI models that could accelerate task generalization
Refuting / qualifying sources:
- Active partner deployments at Mercedes-Benz, GXO Logistics, and Jabil confirmed, but described as “early versions working within designated areas” — scope and productivity metrics not disclosed
- Agility Robotics has more established multi-customer US deployments with less funding and a simpler robot
- Apollo’s 25-DoF complexity is higher than Digit, making reliability validation harder
Summary: Strong institutional backing and credible team. No verified commercial deployment yet. Timeline to commercial scale is uncertain.