Skip to main content

Skild AI

Category: Robotics / Embodied AI

A robotics AI company developing a general-purpose foundation model (the 'Skild Brain') designed to empower diverse robot hardware with human-like adaptability and intelligence. Skild AI was founded in 2023. The company is led by Deepak Pathak. Based in Pittsburgh, USA. Team size: 51-200. Total funding raised: ~$1.83B–$2B+ total (all rounds). Latest round: Series C ($1.4B, Jan 2026 led by SoftBank, Coatue at $14B+ valuation). Key investors include SoftBank Group, NVentures (NVIDIA), Jeff Bezos (Bezos Expeditions), Lightspeed Venture Partners, Sequoia Capital, Coatue, Macquarie Capital, Felicis Ventures, Samsung, LG / LG Technology Ventures, Schneider Electric, CommonSpirit.

AMW Analysis

Skild AI develops a hardware-agnostic foundation model, the Skild Brain, that enables diverse robot platforms to perform tasks without prior training. Founded in 2023 and headquartered in Pittsburgh, the company has raised approximately $1.7 billion in total funding, including a $1.4 billion Series C in January 2026 led by SoftBank and Coatue at a valuation exceeding $14 billion. Key investors include NVIDIA, Jeff Bezos, Lightspeed Venture Partners, and Sequoia Capital.

The company’s recent news flow shows rapid scaling and strategic expansion. In April 2026, Skild acquired Zebra Technologies’ robotics automation unit, adding the Symmetry Fulfillment orchestration platform to integrate its task-learning software across Zebra’s hardware, targeting logistics deployment. The Series C funding, which tripled its valuation in seven months, is being used to scale the Skild Brain, which the company describes as an omni-bodied model trained on vast data. The startup reported $30 million in revenue in late 2025, signaling early commercial traction. The acquisition and funding round reinforce Skild’s position as a provider of a universal robotic intelligence layer, moving toward a software-defined approach to physical automation.

AMW analysis, generated from 8 tracked news signals.

Founded
2023
Headquarters
Pittsburgh, USA
Team size
51-200
Total funding
~$1.83B–$2B+ total (all rounds)

Value proposition

Provides a hardware-agnostic AI 'brain' that allows robots to perform complex tasks in unstructured environments without task-specific programming, significantly reducing deployment costs and time.

Products and solutions

Skild Brain (omni-bodied foundation model for any robot/task), Skild Data Engine / data flywheel (sim, internet video, teleop, real deployments), Hardware-agnostic control / mobile manipulation platform (API), Security & inspection robot applications, Autonomous packing / dexterous manipulation modules, Warehouse robotics stack via acquired Zebra/Fetch platform + Symmetry Fulfillment orchestration

Unique value

Unlike competitors focusing on specific hardware (e.g., humanoids), Skild focuses on a universal intelligence layer that can be deployed across any form factor, from robotic arms to quadrupeds.

Target customer

Robotics OEMs, industrial automation providers, logistics and warehouse operators, and manufacturers seeking flexible automation.

Industries served

Logistics and Warehousing, Manufacturing, Healthcare (Service Robotics), Construction, Consumer/Household Robotics

Technology advantage

Leverages 'Physical AI' trained on a proprietary dataset that is orders of magnitude larger than industry standards, utilizing self-supervised learning techniques to generalize across diverse physical tasks.

How they differentiate

Unlike humanoid-first peers (e.g. Figure), Skild sells an omni-bodied foundation model that controls quadrupeds, humanoids, arms, and mobile manipulators without body-specific retrain-from-scratch. Post-2026 Zebra/Fetch acquisition adds proven warehouse hardware + orchestration, pairing the brain layer with battle-tested logistics deployments and a stronger real-world data flywheel.

Main competitors

Figure AI, Physical Intelligence (π), Sanctuary AI

Key partnerships

NVIDIA / NVentures (Strategic investor and compute partner), Carnegie Mellon University (Research and talent pipeline), Bezos Expeditions (Strategic investor), SoftBank Group (Lead investor across late rounds; strategic capital), Zebra Technologies robotics division / former Fetch Robotics (Acquired Apr 2026 — warehouse robotics + Symmetry Fulfillment platform)

Notable customers

Robotics OEMs, Logistics and Warehouse Operators, Industrial Manufacturing Providers

Major milestones

Emerged from stealth with a $300M Series A in July 2024 at $1.5B valuation, Series B mid-2025 at ~$4.5–4.7B valuation (SoftBank-led), Closed a $1.4B Series C led by SoftBank in January 2026 at $14B+ valuation, Live revenue grew from $0 to ~$30M in a few months of 2025, Acquired Zebra Technologies' robotics division (formerly Fetch Robotics) in April 2026 for warehouse/omni-bodied deployment

Growth metrics

Valuation path: $1.5B (Series A Jul 2024) → ~$4.5–4.7B (Series B mid-2025) → $14B+ (Series C Jan 2026). Live revenue ~$30M in 2025 (company disclosure). Tracxn employee count ~99 as of May 2026 (fits 51-200 band). Acquired Zebra/Fetch robotics arm Apr 2026 to scale warehouse deployments.

Market positioning

Leading provider of general-purpose 'Physical AI' foundation models, positioned as the universal intelligence layer for the global robotics industry.

Geographic focus

North America (Headquartered in Pittsburgh), with a global expansion focus in Asia and Europe through strategic partnerships with SoftBank Group.

Patents and IP

Founders hold extensive intellectual property through Carnegie Mellon University (CMU) related to 'Curiosity-driven learning' and 'Self-supervised robot learning' (e.g., US Patent 11,235,463).

About Deepak Pathak

Dr. Deepak Pathak is the Co-founder and CEO of Skild AI and an Assistant Professor at Carnegie Mellon University's Robotics Institute. He is a renowned researcher in the field of 'Physical AI' and robot learning. His professional background includes serving as a Research Scientist at Meta AI (FAIR) and a researcher at OpenAI. He holds a PhD in Computer Science from UC Berkeley, where he was advised by Trevor Darrell and Alexei Efros, and a BTech from IIT Kanpur. His work focuses on developing general-purpose AI 'brains' that allow robots to adapt to diverse environments and tasks without task-specific programming.

Latest news about Skild AI

More Robotics / Embodied AI companies

Official website: