Neurala offers VIA, a vision-AI software platform built on its Lifelong-DNN (L-DNN) technology to reduce the time, cost and skills required to build and maintain production-quality custom computer-vision solutions. The product is positioned for in-line quality inspection, enabling faster, more accurate and scalable defect detection on manufacturing production lines. Over the last year the company expanded work with system integrators threefold and broadened engagements with OEMs to increase deployment capacity. Neurala launched a European subsidiary in Trieste, Italy to support its growing international footprint while remaining headquartered in Boston, MA. The company says the new capital will be used to evolve VIA for a wider range of applications and accelerate global adoption as manufacturers prioritize automation and Industry 4.0 initiatives. Total invested in Neurala now stands at $26 million. Neurala builds vision AI software focused on industrial and real-world applications, emphasizing edge deployment. Its Lifelong-DNN (L-DNN) technology lowers data requirements and enables continuous learning in the cloud or at the edge. The company helps industrial customers improve quality inspection and monitoring, with deployments across manufacturing, drones, robotics, and smart devices. Neurala’s edge-trained models are designed to extract actionable insights from IIoT sensor and camera data and to be set up by operators without specialized expertise. The company positions its technology as a core enabler of Industry 4.0 initiatives through on-machine, edge AI. Neurala builds the Neurala Brain SDK (marketed as Brains for Bots) to run machine learning locally on resource-constrained devices, enabling incremental learning and on-device updates without continuous network connectivity. The platform is hardware-agnostic, scales from low-end CPUs to high-end GPUs, and integrates with frameworks such as TensorFlow; the SDK supports Linux, Android and iOS. For customers needing custom models, Neurala offers a data ingestion and training service, and it also provides a server-based solution, yielding three complementary business models. The company targets time-sensitive and offline use cases (for example, drones inspecting power lines) and aims to put AI in the hands of toy makers, drone builders and IoT engineers. Early customers include Teal Drones, and Motorola Solutions is noted as an early customer; the founding team and leadership (CEO Massimiliano Versace) have cognitive and neural systems Ph.D.s from Boston University. The article does not disclose revenue or other operating metrics. Neurala builds deep-learning software that enables robots to learn and be instructed by people rather than be explicitly programmed. Its core technology applies artificial neural networks (ANNs) running on GPUs, and the company holds a fundamental patent for that approach. Led by CEO and co-founder Massamiliano Versace, Neurala has secured contracts with NASA and the U.S. Air Force Research Labs to develop deep learning controllers. Use cases cited include autonomous exploration of Mars-like environments, collision avoidance for unmanned aircraft, and persistent surveillance. The company will use the recently raised seed funds to develop practical commercial applications for both flying and ground-based robots that utilize its learning software. Neurala is based in Boston, MA.

Total raised

$27M

Funding rounds

4

Latest round

Equity

Latest activity

Jun 2021