Benefits of its platform include global data access (anywhere, anytime), shorter training times, and up to 10x lower storage costs.

The company is building a software layer to optimize the data flow from edge to cloud from ingestion, filtering, storage, transport, and analytics. Akridata’s software enables the automatically filtering and preparing of the raw data coming from edge sensors to send only the most relevant data points to the cloud, to optimize and accelerate the training of deep learning models. As the volume of data captured by sensors is exploding (while bandwidth remains constant), it is no longer possible or efficient to send all this raw data in the cloud to train AI models. Therefore, two things must happen to prevent costly traffic congestion: moving a chunk of the compute directly at the edge and streamlining the data flows between edge data and cloud to allow faster development of better AI algorithms.