xPU Specialist - Zaspar Technologies

Job Details

Job Title: xPU Specialist

Location: Palo Alto, CA

Type: Full Time

Our client 8bit.AI is a dynamic startup in the Bay Area, CA seeking to hire Full-time employees and focused on developing a high-performance, multi-technology, vendor-independent, xPU-based Accelerated Cloud Computing platform. We stack massive clusters purpose-built for high-performance parallel computing and aim to launch a global accelerated cloud solution. Additionally, the firm will focus on broader Artificial General Intelligence (AGI) products, supercomputing services, and end-to-end AI engineering services.

RESPONSIBILITIES

• Design and implement innovative hardware solutions for highly scalable and efficient xPU PODs.
• Collaborate with architects, software engineers, and system engineers to
• ensure optimal integration of hardware and software components within the PODs
• Deeply understand leading xPU architectures from NVIDIA, AMD, and/or Intel and leverage their capabilities for performance optimization within PODs
• Conduct performance and power analysis to identify and implement strategies for optimizing resource utilization and power consumption within the PODs
• Participate in the development and execution of hardware verification and validation plans
• Stay up to date on the latest advancements in xPU technology and related hardware trends
• Contribute to technical documentation and maintain clear communication within the team

QUALIFICATIONS:

• Master's degree in Electrical Engineering, Computer Engineering, or a related field (or equivalent experience).
• Minimum 8+ years of experience in designing and developing hardware solutions, preferably for data center or high-performance computing environments
• Proven experience with virtualization platforms, preferably including VMware (vSphere, ESXi, etc.) OR Nutanix (AHV, AOS, etc.)
• Strong understanding of hypervisor technologies and their functionalities
• Ability to integrate and manage both internal and external virtualization platforms
• Must have experience developing and running applications using the ROCm platform with a strong understanding of ROCm components like HIP, OpenCL, and AMD GPU architecture
• In-depth knowledge of xPU architectures, particularly from NVIDIA, AMD, or Intel.
• Completed certifications in NVIDIA AI in Datacenter and InfiniBand OR C-DAC certification.
• Strong understanding of computer architecture, memory systems, and interfacing techniques.
• Solid understanding of OpenStack concepts and experience managing cloud infrastructure
• Prior experience in building and operating Cloud POD infrastructure
• Proficiency in hardware description languages (HDL) like Verilog or VHDL
• Experience with hardware simulation and verification tools
• Excellent communication, collaboration, and problem-solving skills
• A passion for innovation and a drive to contribute to cutting-edge technology development

GOOD TO HAVE:

• Experience with hardware design for data center infrastructure or high-performance computing systems
• Experience with thermal and power management solutions for high-density computing environments
• Familiarity with xPU programming frameworks like CUDA or OpenCL.