Shen Sha, senior R&D manager of Kuxin AI chip department, said: "Our customers are pushing forward the innovation of AI application products every day. They urgently need to get the hardware they need to develop software."
Since 2015, Kuxin has been focusing on Image Signal Processor (ISP) and Neural Processing Unit (NPU), these advanced Application-Specific Integrated Chips (ASIC) must meet stringent requirements such as low power consumption and high performance. They represent the most advanced generation of controllers today, using AI for tasks such as object detection, object classification, and object tracking. Kuxin's chips are at the heart of the latest generation of drones, and are also used in areas such as intelligent surveillance systems and driverless cars.
The design of these chips has the characteristics of large scale and high complexity, and the verification of the wafer design is also a big challenge. Shen Sha said: "We use S2C's FPGA prototype verification system on a large scale, including their Prodigy core VU440 and UV19P single-core and dual-core systems, so we can quickly and effectively complete the verification of 100 million-gate complex chips."
But validating hardware still requires testing and validating millions of lines of code. As a company that focuses on customer experience, Kuxin strives to find new solutions. Shen Sha said: "In terms of specific needs, we used the S2CFPGA prototyping solution to directly build prototypes for customers to evaluate. This really accelerated their software development process."
Sharing S2C's Prodigy VU440 prototyping platform enables the two companies to form a unique form of collaboration -- Kuxin can rapidly verify 100 million gate designs and help customers acquire the necessary hardware to advance development plans. In addition, by working closely with customers, Coolcore has gained deeper insights and knowledge that can help drive improvements in its own designs. This is a win-win strategy for both companies.
S2C specializes in providing leading prototyping platforms for designs and applications of all sizes. Built on the world's largest and fastest FPGAs, such as the Xilinx UltraScale+ VU19P and Intel Stratix 10, the company's Prodigy family can scale up to meet the verification needs of very large designs. With an extensive library of daughter cards and interface modules, the Prodigy system can be quickly adapted for any application. If you are challenged with large-scale design verification, we do our best to find the most efficient solution for you.
is the leading embedded AI SoC supplier. The company's chips are widely used in intelligent security, intelligent hardware, intelligent vehicle and other fields. Kuxin has accumulated rich experience in wireless communication, computer vision and deep learning, and has established a high reputation in the industry.