AMD Expands AI Capabilities with Acquisition of Enosemi Team
AMD has announced the acquisition of the Enosemi team, a key step in the company’s expanding AI strategy. Having previously collaborated with AMD as an external development partner in photonics, Enosemi will now join AMD, helping to scale the company’s ability to develop cutting-edge photonics and co-packaged optics solutions for next-gen AI systems.
A Team of Experts in Photonics
Enosemi, based in Silicon Valley, is renowned for its expertise in photonic integrated circuits (PICs). The team’s experience includes building and shipping PICs in volume, a feat accomplished by only a select few in the industry. Their proven track record and technical excellence make them a perfect fit for AMD as the company accelerates its high-performance interconnect innovations, which are increasingly crucial as AI workloads become more complex and demanding.
Strengthening AMD’s AI Portfolio
AMD’s acquisition of Enosemi complements its broader AI strategy, which has seen the company build a best-in-class portfolio aimed at addressing the rapidly evolving needs of AI.
This includes integrating Xilinx’s AI Engines, adding networking capabilities through Pensando, and enhancing software with the acquisitions of Silo AI and Mipsology. Furthermore, AMD has scaled full rack-level system design with the acquisition of ZT Systems.
As AI models grow in size and complexity, the need for faster and more efficient data movement is becoming increasingly important. To address this, optical interconnects, such as co-packaged optics, are emerging as a promising solution.
These advanced interconnects offer higher bandwidth density and greater power efficiency compared to traditional methods, marking a transformative step in system architecture. The integration of compute and networking at the rack scale will be essential for supporting the performance and scale required by next-generation AI workloads.
Also read: Humain and AMD Announce $10 Billion AI Infrastructure Push
Full-Stack Innovation for AI Systems
Looking ahead, AMD recognizes that the demands of AI systems will require more than just powerful chips. End-to-end innovation across compute, networking, systems architecture, and software is needed in order to address these changing needs. With its leading-edge CPUs, GPUs, adaptive SoCs, and now bolstered networking and system integration technologies, AMD is well-placed to make this vision a reality.
