Mecka Secures $60M to Scale Real-World Robotics Data and Deployment Systems
Robotics infrastructure startup Mecka has raised $60M in Series B funding led by Sequoia Capital to expand its physical AI data capture and integration services.
Robotics infrastructure startup Mecka has secured $60 million in a Series B funding round led by Sequoia Capital. The fresh capital will be used to expand the company's real-world robotics data collection systems, accelerate its research initiatives, and scale its commercial robot deployments. A diverse group of prominent technology companies and venture funds participated in the round, including Nvidia, Microsoft's venture fund M12, Qualcomm Ventures, Samsung, Kindred, Framework Ventures, and Neo. Individual backers include DoorDash founder and CEO Tony Xu, former ServiceNow and Snowflake CEO Frank Slootman, and Milan Kovac, the former head of Tesla's Optimus program.
Bridging the Real-World Data Gap
Mecka's primary focus is addressing a critical bottleneck in the robotics industry: the lack of high-quality physical training data. While traditional AI models leverage vast amounts of text and images harvested from the internet, physical robots require specialized data to perform complex manual tasks. Common actions such as folding, pouring, or grasping involve motion, force, contact, and three-dimensional positioning—data that is rarely captured in digital-only formats.
To solve this, Mecka has built its own end-to-end data pipeline. The company designs and manufactures proprietary multi-sensor hardware to record human demonstrations in real-world settings, such as homes and commercial spaces. This physical signal data is then processed using Mecka's computer-vision and multimodal models, which handle motion tracking, 3D reconstruction, and sensor alignment. According to the company, this setup achieves sub-centimeter hand-pose tracking accuracy even outside controlled laboratory environments.
In addition to its proprietary work, Mecka collaborates with academic and corporate research institutions. The company has partnered with researchers at Stanford, MIT, Georgia Tech, UC San Diego, ETH Zürich, and Meta on EgoVerse, a study exploring how human demonstrations can be effectively transferred to different robots and tasks across various environments.
Moving Beyond Data Collection to Full Integration
Mecka is positioning itself as more than just a data provider. The company is actively expanding its business model to act as a full-service robotics integrator for enterprises that lack internal robotics expertise. Under this model, Mecka provides the physical robot hardware, records demonstration data directly at the customer's site, performs post-training on the systems, and manages ongoing operations. This continuous loop allows deployed robots to steadily improve as they gather more real-world operational data over time.
This comprehensive approach has fueled rapid financial growth. Mecka reported surpassing a $100 million run-rate revenue in June 2026, just months after starting operations. The company projects its run rate will reach $300 million by the end of 2026 as it scales its integration services and expands its customer base, which already includes several major robotics research laboratories and multiple "Mag 7" technology companies.
What it means for developers
For developers working at the intersection of software and physical AI, Mecka's progress highlights a shifting paradigm where data collection and post-training are becoming standardized services. Instead of building custom data-capture rigs or managing complex sensor calibration pipelines from scratch, developers can increasingly rely on unified infrastructure to feed physical demonstration data directly into their neural networks.
Furthermore, the rise of physical AI requires robust cognitive architectures to interpret environment states and plan complex actions. Developers building these high-level decision-making systems can try top AI models cheaply through one API at https://apixoai.online. This single-key access allows teams to easily benchmark different LLMs and multimodal models for robot task planning, semantic understanding, and natural language interfaces without managing multiple vendor relationships.
As Mecka’s ecosystem expands to accommodate diverse robot form factors and tasks, the barrier to entry for deploying physical automation will continue to fall, allowing software developers to focus on high-level logic and application design rather than hardware integration.
Source: Mecka Raises $60M in Series B Funding to Build Data and Deployment Infrastructure for Robots — AI Insider. Written by the Apixo team from that report.
One key for Claude, GPT, GLM, DeepSeek and more. Pay per token with crypto.
Get your API key
