* The modular AI components could speed up development.
* the DeepScale software could make self-driving more energy efficient

In January, 2019, DeepScale introduced Carver21 which is AI building blocks for intelligent cars. Carver21 offers a portfolio of perception software modules that give Tier-1s and OEMs the flexibility they need to create modular, scalable ADAS. (advanced driver-assistance systems).

Carver21 can be used in small SoCs that would be embedded in edge ECUs, which typically require less than 5W power consumption and are still the status quo for the distributed electrical architectures used in today's ADAS.

DeepScale's full-stack deep learning methodology enables cohesive integration of AI software with various processors and sensors for customizable automated driving features. Full-stack deep learning means that DeepScale has experts working together on every aspect of deep neural network (DNN) training, development, deployment, and even data collection/curation to produce proprietary state-of-the-art AI solutions.

DeepScale can run three parallel deep neural networks while using only ~2% of Nvidia's Drive AGX Xavier processing. Xavier has 30 TOPS (trillion operations per second) of available power.