Humyn Labs Raises $47M to Scale Physical AI Data

Humyn Labs Raises $47M to Scale Physical AI Data Humyn Labs Raises $47M to Scale Physical AI Data

Humyn Labs has raised $47 million in Series A funding to expand its real-world robotics data collection, enrichment and deployment-readiness infrastructure across Brazil, South Korea and India. The round combines $26 million in equity led by Accel and $21 million in venture debt led by Lighthouse Canton, as the company targets the data and deployment infrastructure required to bring physical AI systems into real-world environments.

Humyn Labs Raises $47M to Expand Physical AI Infrastructure

The development of physical AI models is increasing demand for real-world data that captures how people perform tasks across industrial, commercial and everyday environments. Humyn Labs is positioning itself in that emerging infrastructure market, using human demonstrations and data enrichment workflows to help robotics developers train and deploy autonomous systems.

The company announced $47 million in Series A funding on Nov. 19, 2026. The financing consists of a $26 million equity round led by Accel and a $21 million venture debt round led by Lighthouse Canton.

Equity investors include Prosus Ventures, BeeGlobal, Finserv Intelligence through Bajaj Finance, Bajaj Alts AI Opportunities Fund, Dallas Venture Capital, Lighthouse Canton, Jump Trading LLC and angel investor Anita Pramoda. Trifecta Capital participated in the venture debt financing.

Humyn Labs says it surpassed $50 million in annualized revenue within six months of operations. The company works with frontier AI labs, robotics companies and world-model developers, operating data capture and enrichment programs across more than 20 countries.

Turning human work into robotics training data

Humyn Labs focuses on capturing human activity and translating it into structured datasets for physical AI models. Its operations span factories, warehouses, farms and homes, where workers perform tasks that robots may eventually need to replicate or assist with.

The company uses proprietary capture devices to record specific activities according to requirements provided by AI research laboratories and robotics developers. Its data engine then converts the recordings into labeled, training-ready datasets with verified provenance across sight, sound, motion and touch.

The approach addresses a challenge facing physical AI development: models need more than conventional internet data to learn how objects behave, how tasks are performed and how machines should respond to physical environments.

Robotics training may require demonstrations involving object manipulation, movement, tool usage, workplace procedures and interactions with different materials or equipment. Capturing these activities in structured formats can help developers build datasets for training and evaluating AI systems designed to operate in physical settings.

Humyn Labs is positioning itself as an infrastructure provider between real-world environments and AI research organizations, supplying the data capture and enrichment processes required to support robotics development.

Physical AI data requires real-world context

Physical AI models must operate in environments where objects, conditions and workflows can change. Unlike digital AI tasks, robotics applications involve physical constraints, spatial relationships, movement, contact and the consequences of incorrect actions.

This creates demand for datasets that capture the complexity of human work. Video alone may not adequately represent how a person manipulates an object, applies force, handles tools or adapts to unexpected conditions. Combining visual, auditory, motion and touch-related information could help researchers develop richer representations of real-world tasks.

Humyn Labs’ data capture and enrichment model is designed to collect such information and make it usable for AI research. The company says it works directly with model labs and site operators, aligning capture operations with specific training requirements.

The strategy places data infrastructure alongside robotics hardware, simulation and model development as a component of the broader physical AI ecosystem. Companies developing embodied AI systems increasingly need data pipelines that connect research environments with practical operating conditions.

The effectiveness of this approach will depend on dataset quality, task diversity, labeling accuracy, provenance, privacy safeguards and how well the collected data translates into robot performance.

Funding supports deployment-readiness expansion

Humyn Labs plans to use the capital to expand deployment-readiness infrastructure in Brazil, South Korea and India, strengthen its capture hardware and data enrichment engines, and establish a broader operational presence in San Francisco, London and Seoul.

The geographic expansion is intended to support its work with robotics companies, AI labs and world-model developers across multiple markets.

Founded in 2026 by Manish Agarwal, an alumnus of IIM Ahmedabad, and Ishank Gupta, an alumnus of The Wharton School, the company is positioning its operations around the relationship between human expertise and machine learning.

The founders argue that robots need to learn from workers performing real tasks in the environments where those tasks occur, rather than relying exclusively on online datasets or simulated demonstrations.

The funding also reflects investor interest in the infrastructure surrounding physical AI, including data collection, training, deployment, simulation and hardware integration.

Physical AI infrastructure draws investor attention

Physical AI is expanding beyond robotics hardware into a broader technology stack encompassing AI models, data pipelines, simulation environments, sensing, compute infrastructure and deployment services.

Humyn Labs is targeting the data layer of that stack. Its model focuses on gathering human demonstrations and converting them into structured training material that can be used by robotics developers.

The competitive ecosystem includes AI labs developing embodied models, robotics manufacturers building autonomous platforms and companies creating simulation and training infrastructure. Organizations such as NVIDIA and Google DeepMind are investing in robotics research, simulation and embodied intelligence, while robotics startups are working to bring autonomous systems into industrial and commercial settings.

Humyn Labs’ reported annualized revenue of more than $50 million within six months of operations, if sustained, would indicate early commercial demand for its data and deployment services. However, annualized revenue is not the same as audited annual revenue, and the company has not disclosed the composition of its revenue, customer concentration or profitability.

The next challenge will be scaling data capture and enrichment while maintaining quality, consistency and operational efficiency across different industries and countries.

As robotics developers move toward real-world deployments, the ability to capture and structure human experience could become an important component of physical AI development. Humyn Labs is betting that data infrastructure will be a foundational part of that transition.

Market Landscape

Physical AI is creating demand for infrastructure that extends beyond AI model development and robotics hardware. Training autonomous machines requires real-world data, simulation, model evaluation, hardware integration and deployment infrastructure.

Humyn Labs is targeting the real-world data layer of this ecosystem.

Key market trends include:

  • Physical AI data collection: Human demonstrations provide contextual information about tasks, objects, movement and operational environments.
  • Robotics training datasets: Structured, labeled and verified data supports model training, testing and evaluation.
  • Embodied AI: Models are being developed to understand physical environments and translate intelligence into actions.
  • Robotics deployment infrastructure: AI developers increasingly need real-world testing, integration and deployment support beyond model development.
  • Human-to-robot learning: Capturing how experienced workers perform tasks could help robotics developers build more capable autonomous systems.

The broader competitive landscape includes robotics data providers, simulation companies, embodied AI research labs, foundation model developers and autonomous hardware manufacturers. Humyn Labs is positioning its capture hardware and data enrichment operations as infrastructure that connects real-world work with AI model development.

Top Insights

  • Humyn Labs raised $47 million in Series A financing, comprising $26 million in equity and $21 million in venture debt.
  • The company says it surpassed $50 million in annualized revenue within six months of operations.
  • Its data infrastructure captures human work across factories, warehouses, farms and homes for physical AI training.
  • Humyn Labs plans to expand deployment-readiness infrastructure in Brazil, South Korea and India.
  • The company is targeting the physical AI data infrastructure market through capture hardware, dataset enrichment and verified data provenance.

Power Tomorrow’s Intelligence — Build It with TechEdgeAI

Grow Your
Brand Visibility

Looking to publish a press release, guest article, interview or podcast? Connect with us.

GET FEATURED
Subscribe

Sign up today for exclusive insights and updates.

Newsletter Signup