Chenglong Zeng

Title: Chenglong Zeng: Innovator in AI Accelerator Technology

Introduction

Chenglong Zeng is a prominent inventor based in Guangdong, China. He has made significant contributions to the field of artificial intelligence, particularly in the development of storage and accessing methods for parameters in streaming AI accelerator chips. With a total of 3 patents to his name, Zeng is recognized for his innovative approaches to enhancing AI technology.

Latest Patents

Zeng's latest patents include groundbreaking technologies that improve the efficiency of AI accelerators. One of his notable inventions is a method for storage and accessing parameters in a streaming AI accelerator chip. This technology involves a streaming-based data buffer that comprises multiple banks, each configured to store different data. The data read circuit is designed to read first data required for computation tasks in parallel from these banks, significantly enhancing processing speed.

Another important patent is the data stream architecture-based accelerator, which includes a storage unit, a read-write address generation unit, and a computing unit. This architecture allows for efficient data processing by determining target banks and data paths based on preset parallelism, ensuring that operations on data are performed swiftly and effectively.

Career Highlights

Chenglong Zeng is currently employed at Shenzhen Corerain Technologies Co., Ltd., where he continues to push the boundaries of AI technology. His work has positioned him as a key player in the development of advanced AI systems, contributing to the company's reputation as a leader in the field.

Collaborations

Zeng collaborates with talented individuals such as Kuen Hung Tsoi and Xinyu Niu, who share his passion for innovation in technology. Together, they work on projects that aim to revolutionize the capabilities of AI accelerators.

Conclusion

Chenglong Zeng is a visionary inventor whose work in AI accelerator technology is paving the way for future advancements in artificial intelligence. His patents reflect a deep understanding of the technical challenges in the field and a commitment to finding innovative solutions.