Startup Expands Deep Learning CPU Development
An Israeli chip startup developing a new processor architecture for embedded AI applications has lined up another group of investors as it seeks to add intelligence to embedded devices at the network edge.
Tel Aviv-based Hailo announced this week it raised $12 million in a Series A funding round that included Ourcrowd.com, Maniv Mobility, NextGear Ventures and angel investors. The startup has so far raised $16 million.
The company said it would use the new funding to expand development of its deep learning processor designed to deliver “datacenter processing capacity” to edge devices. Hailo expects to release initial samples of its processor during first half of 2019. Among the embedded AI applications are edge devices installed on autonomous vehicles, drones and other gadgets linked via the Internet of Things and other emerging platforms.
Hailo’s deep learning architecture is billed as nothing less than a reworking of the memory, control and compute as well as the “relations between them,” the startup said. The deep learning processor seeks to significantly reduce size, power consumption and price to provide local processing of sensor data in real time.
“Existing general-purpose processor infrastructure cannot efficiently run compute-intensive deep learning algorithms necessary for these applications,” the startup noted in announcing the funding round on Tuesday (June 5).
The startup and its investors are betting the nascent market for autonomous vehicles will continue to drive demand for AI building blocks that will require deep learning capabilities. Similarly, leading chip makers like Intel (NASDAQ: INTC), which is combining its Xeon scalable CPUs with FPGA technology, and GPU specialists such as Nvidia (NASDAQ: NVDA) are eyeing the autonomous vehicle market as demand for model training and inference expands.
Last year, Intel acquired another Israeli startup, computer vision specialist Mobileye, to pursue automotive applications like collision avoidance.
Hailo’s leadership includes CEO Orr Danon, who previously served in the Israeli Defense Force’s technology unit, and CTO Avi Baum, a former senior engineer at Texas Instruments (NASDAQ: TXN). Its development team includes hardware and software engineers who worked at Intel, Broadcom (NASDAQ: AVGO) and Mellanox (NASDAQ: MLNX).
The startup also said it is collaborating with unnamed industry partners on the development of its deep learning processor.