Company attributes
Other attributes
Edge Impulse is a developer of artificial intelligence and machine learning technology to enable intelligent device solutions and edge compute solutions. This includes data collection, model building, deployment, and monitoring systems. The company's platform can be used to contribute to and extend both device support and algorithms to help developers build new intelligent devices for new applications. This includes for edge compute applications for agriculture, appliances, buildings, the environment, health, infrastructure, and wearables.
Edge Impulse was founded in 2019 by CEO Zach Shelby and CTO Jan Jongboom and is headquartered in San Jose, California. The company was founded to help developers develop new intelligent devices that can have a positive impact on society with a specialization in industrial and professional applications, including predictive maintenance, asset tracking and monitoring, and human and animal sensing. The company has partnerships with Arm, Hackster, TensorFlow, Tiny ML, ETA Compute, and Arduino. Customers of Edge Impulse have included Oura, Nordic Semiconductor, Poly, NASA, Sony, and Bosch.
The Edge Impulse Platform is a development platform for edge machine learning on embedded devices for sensors, audio, and computer vision, among other applications, at scale. The platform enables the deployment of optimized machine learning on hardware ranging from MCUs to CPUs and customer artificial intelligence accelerators.
The platform allows users to build custom datasets with flexible data ingestion from various devices, files, or cloud integrations; offers dataset automation to automatically label data from devices; allows users to set up and run reusable scripted operations that transform input data in parallel through cloud infrastructure; and offers various integrations. The platform allows users to develop models and algorithms, either with pre-built models and algorithms offered by Edge Impulse with various customizations, or through custom development of those models or algorithms.
Further, the platform allows users to test and optimize those models and algorithms for model performance through virtual cloud simulations and live classifications. And the platform is developed to help users deploy those systems and models on any edge target and to optimize source codes and algorithms for those devices. This includes using cloud-hosted machine learning projects and any hardware for digital twin capabilities.
