Skild AI: $14B Robotics Brain Startup (Founded 2023)

Skild AI: Pittsburgh startup building the Skild Brain, the world's first omni-bodied robotics foundation model. Founded 2023, raised $2.2B, valued at $14B+.

Skild AI is a Pittsburgh startup founded in 2023 by Deepak Pathak and Abhinav Gupta, both former Carnegie Mellon professors. The company builds the Skild Brain, a foundation model designed to control any robot body for any task. Skild AI has raised approximately $2.2B across four rounds (most recently a $1.4B Series C in January 2026) and is valued at over $14B.

Skild AI, founded in 2023 by former Carnegie Mellon professors Deepak Pathak and Abhinav Gupta, is a Pittsburgh-based startup building the Skild Brain: the world's first unified robotics foundation model. The Skild Brain can control any robot (humanoids, quadrupeds, robotic arms) without prior knowledge of its body form. As of January 2026, the company has raised approximately $2.2B across four rounds and is valued at over $14B.

Founded: 2023 · HQ: Pittsburgh, PA, USA · Team: 50-100 · CEO: Deepak Pathak · Funding: $2.2B total raised across seed (2023, $14.5M), Series A (July 2024, $300M, led by Lightspeed and Coatue), Series B (summer 2025, ~$500M, led by SoftBank), and Series C (January 2026, $1.4B, led by SoftBank) · Valuation: $14B+ (Series C, January 2026)

About Skild AI

Skild AI was founded in early 2023 by Deepak Pathak and Abhinav Gupta, two computer science professors who left Carnegie Mellon University to pursue a shared thesis: that the robotics industry did not need hundreds of task-specific AI models but a single general-purpose brain capable of driving any robot through any physical challenge. Pathak, who serves as CEO, spent years at CMU studying curiosity-driven learning and self-supervised robot control. Gupta, who serves as President, contributed expertise in large-scale robot learning and simulation-to-real transfer. The pair earned the Best Robotic System Award at the Conference on Robot Learning in 2022 for work on adaptive sim-to-real training. The company is incorporated as a C-corporation with its principal office in Pittsburgh, Pennsylvania, and a secondary engineering hub in Pasadena, California. The company's flagship product is the Skild Brain, which it describes as the world's first unified robotics foundation model. Unlike task-specific systems that require separate training for each robot type, the Skild Brain is designed to be omni-bodied: it can control any robot (humanoid, quadruped, tabletop arm, or autonomous mobile robot) without prior knowledge of that robot's exact mechanical structure. The model pre-trains on two unconventional data sources: publicly available human videos of everyday physical tasks, and procedurally generated physics-based simulations. At inference, the Skild Brain runs real-time perception and motor control simultaneously, adapting to the robot body it operates rather than relying on hard-coded motor primitives or pre-mapped kinematics. Skild AI unveiled the Skild Brain publicly in July 2025, releasing footage of humanoid robots, quadruped dogs, and tabletop manipulation arms all running under a single model. In April 2026, the company announced the acquisition of Zebra Technologies' Robotics Automation Business, including its Symmetry Fulfillment warehouse orchestration platform. This acquisition made Skild AI the first organization offering an end-to-end physical automation stack: humanoids for pick-and-place tasks, quadrupeds for facility inspection, robotic arms for packing, and autonomous mobile robots for material transport, all coordinated by a single software layer. Skild AI is also working with NVIDIA and Foxconn on factory floor deployments, with initial pilots underway in 2026. Skild AI raised a $14.5 million seed round in 2023 to fund initial research and hiring. In July 2024, the company announced a $300 million Series A led by Lightspeed Venture Partners and Coatue, with participation from SoftBank, Bezos Expeditions, Sequoia Capital, and General Catalyst, at a post-money valuation of $1.5 billion. By summer 2025, Skild AI closed an approximately $500 million Series B led by SoftBank, with NVIDIA (via NVentures), Samsung, and LG also investing, at a valuation of roughly $4.5 billion. In January 2026, SoftBank led a third consecutive round, a $1.4 billion Series C, with new participation from Macquarie Capital, Disruptive, and 1789 Capital, alongside returning investors Felicis, Coatue, Lightspeed, and Sequoia. The Series C brought total raised to approximately $2.2 billion and the valuation to over $14 billion, making it the largest robotics AI funding round on record at the time. Skild AI's business model is B2B licensing: robot manufacturers, logistics operators, and industrial enterprises pay to run the Skild Brain on their physical systems. The company does not sell robots directly; it sells the intelligence layer that runs on hardware from partners and customers. Revenue reportedly grew from zero to approximately $30 million in 2025, with deployments spanning security inspection, last-mile delivery, warehouse automation, manufacturing, data center operations, and construction tasks. The company has not publicly disclosed contract sizes or per-robot licensing rates, and no public-facing subscription tiers exist as of mid-2026. Deepal Pathak serves as CEO, responsible for product direction and investor relations. Abhinav Gupta holds the title of President, overseeing research and technical operations. Both founders remain active researchers and are based primarily in Pittsburgh. As of March 2026, Tracxn reported approximately 64 employees, a deliberately lean headcount relative to the $2.2 billion in raised capital, reflecting a strategy centered on high-density research and engineering hires. The company maintains offices in Pittsburgh (primary, robotics and AI research) and Pasadena, California (secondary hub for West Coast engineering talent). Skild AI's stated mission is to build a single, general-purpose artificial brain that can control any robot for any task, which the founders describe as building AGI for the physical world unconstrained by robot type or environment. The core research bet is that scaling cross-embodiment training data, including human video at internet scale and procedurally generated physics simulation, produces a foundation model with enough physical priors to generalize to new robot bodies without task-specific fine-tuning. This is distinct from the prior dominant approach in robotics AI, which required separate datasets and separate models for each robot body, each task, and each environment. The Skild Brain handles both perception (understanding the robot's surroundings via cameras and sensors) and motor control (generating joint commands) in a single unified pass, rather than decomposing these into sequential modules. Research areas include sim-to-real transfer, cross-embodiment learning, emergent locomotion, and manipulation from human video demonstrations. In the robotics AI foundation model space, Skild AI's closest peer is Physical Intelligence (pi.ai), which raised over $900 million by 2025 and focuses on precision manipulation tasks for tabletop and upper-body humanoid robots. Skild AI's differentiator is the omni-bodied claim: one model controlling quadrupeds, humanoids, and mobile manipulators simultaneously without body-specific retraining. Figure AI, backed by OpenAI and Microsoft, targets humanoid robots specifically in automotive assembly with deep OEM integration at BMW and other manufacturers, a narrower but deeper approach. Where Skild AI faces pressure: a team of approximately 60-100 engineers is small relative to the competitive intensity in this space, and annual revenue of roughly $30 million is early for a $14 billion valuation, requiring rapid commercial scaling. Skild AI is a US-incorporated company based in Pittsburgh, which positions it favorably relative to Chinese robotics competitors in CFIUS-relevant reviews for defense-adjacent applications. Foreign investors including SoftBank (Japan), Samsung (Korea), LG (Korea), and Macquarie (Australia) hold positions in the company, which may require periodic CFIUS filings as the company pursues government or defense-adjacent contracts. GDPR obligations apply to any EU-resident personal data processed through enterprise contracts. No EU AI Act classification has been publicly disclosed, and no FedRAMP authorization has been sought as of mid-2026. The April 2026 acquisition of Zebra Technologies' Robotics Automation Business marks Skild AI's shift from pure-software licensing to vertically integrated physical automation with an orchestration layer. Near-term priorities include expanding the Skild Brain to additional robot body types, deepening the NVIDIA and Foxconn factory deployments, and scaling revenues to support the $14 billion valuation. Analyst estimates place an IPO at least three years away, as the company is in early commercial scale rather than mature revenue territory. The next 12 months will test whether the omni-bodied foundation model approach converts into durable enterprise contracts across the warehouse, healthcare, and construction sectors.

Mission

Building a single, general-purpose artificial brain that can control any robot for any task.

Products

Links

Website · Twitter · LinkedIn · Blog

Frequently Asked Questions

What is Skild AI and what do they build?

Skild AI is a Pittsburgh-based startup, founded in 2023, that builds the Skild Brain, billed as the world's first unified robotics foundation model. The company was founded by Deepak Pathak and Abhinav Gupta, both former Carnegie Mellon University professors, with the goal of creating a single AI system capable of controlling any robot for any task. Unlike traditional robotics AI systems that require separate models for each robot type and task, the Skild Brain uses an omni-bodied approach: one model controls humanoids, quadrupeds, tabletop arms, and autonomous mobile robots interchangeably. The model pre-trains on human videos of everyday physical tasks and physics-based simulations rather than robot-specific teleoperation data, which allows it to generalize to new robot bodies without retraining. By April 2026, Skild AI had also acquired Zebra Technologies' Robotics Automation Business, adding the Symmetry Fulfillment warehouse orchestration platform to its product stack. The company serves customers in warehouse automation, manufacturing, facility inspection, and data center operations, and reported revenue of approximately $30 million in 2025.

Who founded Skild AI and who is the CEO?

Skild AI was co-founded in 2023 by Deepak Pathak (CEO) and Abhinav Gupta (President), both former tenured professors in the School of Computer Science at Carnegie Mellon University. Pathak's prior research focused on curiosity-driven learning, self-supervised robot control, and exploration in reinforcement learning, areas he published on extensively before founding the company. Gupta contributed expertise in large-scale robot learning, sim-to-real transfer, and adaptive control; the pair jointly won the Best Robotic System Award at the Conference on Robot Learning in 2022. The founding thesis was that robotics AI had remained fragmented because data collection was too narrow and robot-specific; Pathak and Gupta left CMU to build the cross-embodiment foundation model they believed the field needed. Deepak Pathak serves as CEO, leading product strategy and investor relations; Abhinav Gupta serves as President, overseeing research and technical operations. Both co-founders remain the primary public faces of the company as of mid-2026, with no other executive names publicly disclosed.

How much funding has Skild AI raised?

Skild AI has raised approximately $2.2 billion across four funding rounds since founding in 2023. The company started with a $14.5 million seed round in 2023 to fund initial research and hiring. In July 2024, Skild AI announced a $300 million Series A led by Lightspeed Venture Partners and Coatue, with participation from SoftBank, Bezos Expeditions, Sequoia Capital, and General Catalyst, at a post-money valuation of $1.5 billion. By summer 2025, the company closed an approximately $500 million Series B led by SoftBank, with NVIDIA (via NVentures), Samsung, and LG also investing, at a valuation of roughly $4.5 billion. In January 2026, SoftBank led a third consecutive round, a $1.4 billion Series C with participation from Macquarie Capital, Disruptive, 1789 Capital, Felicis, Coatue, Lightspeed, and Sequoia, bringing the valuation to over $14 billion and making it the largest robotics AI funding round on record. The company is privately held with no IPO planned for at least three years.

What products does Skild AI make?

Skild AI's primary product is the Skild Brain, a robotics foundation model designed to act as the intelligence layer for any type of robot, including humanoid robots, quadruped dogs, tabletop manipulator arms, and autonomous mobile robots. The Skild Brain is licensed to robot manufacturers, logistics operators, and industrial enterprises on a B2B basis; no consumer product or public API exists as of mid-2026. The model was publicly unveiled in July 2025, with demonstration footage showing a single model controlling fundamentally different robot body types without separate training runs. In April 2026, Skild AI acquired Zebra Technologies' Robotics Automation Business, bringing the Symmetry Fulfillment orchestration platform into the company's product portfolio to coordinate multiple robot types in warehouse environments. Skild AI also has active technology partnerships with NVIDIA (using Omniverse for physics simulation) and Hewlett Packard Enterprise (HPE AI compute for training workloads). No public pricing tiers or per-license fees have been disclosed.

Where is Skild AI headquartered and how big is the team?

Skild AI is headquartered in Pittsburgh, Pennsylvania, the same city where both co-founders built their academic careers at Carnegie Mellon University. The company also operates a secondary engineering hub in Pasadena, California, to access West Coast robotics and AI talent. As of March 2026, Tracxn reported approximately 64 employees, a deliberately lean headcount relative to the $2.2 billion in raised capital, reflecting a strategy of hiring high-density research and engineering talent rather than scaling broad functions early. The team grew from a handful of researchers in 2023 to roughly 50-100 people by early 2026, with headcount concentrated in machine learning research, robotics engineering, and enterprise sales. Both co-founders remain based in Pittsburgh and active as researchers, keeping the company's research culture closely tied to its academic roots. No major office expansions or layoffs have been disclosed as of mid-2026.

What is Skild AI's mission or research focus?

Skild AI's stated mission is to build a single, general-purpose artificial brain that can control any robot for any task, a goal the founders describe as building AGI for the physical world. The company's core research bet is that scaling cross-embodiment training data, including human video from the internet and procedurally generated physics simulations, produces a foundation model with enough physical priors to generalize to new robot bodies without task-specific fine-tuning. This is distinct from the dominant prior approach in robotics AI, which required separate datasets and separate models for each robot body, each task, and each environment. The Skild Brain's architecture handles both perception (understanding the robot's environment via cameras and sensors) and motor control (generating joint commands) simultaneously in real time, rather than decomposing these into sequential modules. Research areas include sim-to-real transfer, cross-embodiment learning, emergent locomotion, and manipulation learned from human video demonstrations. The company maintains informal research ties to Carnegie Mellon through both co-founders' prior academic networks.

Is Skild AI compliant with SOC 2, GDPR, HIPAA?

Skild AI has not publicly disclosed any SOC 2, ISO 27001, HIPAA, or GDPR certifications as of mid-2026, which is common for a three-year-old B2B startup at this stage of commercial scale. The company does not operate a publicly accessible trust center or security page where enterprise compliance documentation would normally appear. Customers evaluating Skild AI for sensitive deployments in defense logistics or healthcare robotics should request a vendor security questionnaire directly from the company's enterprise sales team. GDPR obligations apply to any EU-resident personal data Skild AI processes through enterprise contracts, though the primary product (robot motor control) does not inherently involve personal data collection. No HIPAA Business Associate Agreement availability has been disclosed, and no FedRAMP authorization has been sought as of mid-2026. Compliance posture is expected to mature as the company scales enterprise sales, particularly after the Zebra Technologies acquisition introduces more regulated-industry customers.

Who are Skild AI's main competitors?

Skild AI competes most directly with Physical Intelligence (pi.ai), a San Francisco startup that raised over $900 million through 2025 and focuses on precision manipulation in tabletop and upper-body humanoid contexts; Physical Intelligence has a deeper focus on fine-grained dexterity while Skild AI claims broader cross-robot generalization. Figure AI, backed by OpenAI, Microsoft, and Bezos Expeditions, builds humanoid robots specifically for automotive assembly lines (BMW among others), taking a vertically integrated approach that is narrower in robot type but deeper in OEM partnership. 1X Technologies, backed by OpenAI, and Apptronik, partnered with NASA and Goldman Sachs, also build humanoid robots but target specific embodiments rather than a cross-robot foundation model, placing them in adjacent rather than direct competition. Boston Dynamics (owned by Hyundai) competes on physical hardware robustness rather than general AI software, making it less of a direct competitor on the model layer. Skild AI's stated advantage over all of these is the omni-bodied claim: one model running across quadrupeds, humanoids, and mobile manipulators simultaneously without retraining. The company's risk is that the foundation model approach, if validated at scale, may be replicated by well-resourced players including NVIDIA or Google DeepMind.