System Engineer, AI Accelerator Module Management

The Accelerator Reference Design Team is looking for a System Engineer to design, implement, and support hardware management for custom AI hardware. The ARD team is working on the design and implementation of hardware modules for Meta's custom AI accelerators. Meta is developing large-scale AI/HPC clusters using custom-designed AI accelerators. In this role, you will have a unique opportunity to shape the future AI/HPC of Meta by specifying technical requirements for device management, driving specifications and designs for integrating device management into fleet management, and steering the industry and ecosystem partner direction. The ideal candidate will work in a cross-functional engineering environment, prioritizing competing workstreams based on impact, deadlines, and stakeholder needs. They will have experience in solving complex device management problems in high-performance computing that span across silicon, hardware, and software. They will also have direct experience in hardware design and management of complex ASICs. A successful candidate will have experience working on hardware management across rack, server, and ASIC levels. The candidate will have experience with current industry practices and standards for device management and debug. The position requires a lead developer to architect solutions, debug complex issues, define scope for open-ended technical challenges, and deliver hardware management at scale. The candidate will work closely with cross-functional stakeholders and partners who are on the front-line of developing Meta's custom ASICs for AI. The Accelerator Reference Design Team designs, builds, brings-up, tests and integrates hardware systems that power Meta's custom AI silicon platforms, deployed in data centers worldwide. In this role, you will help design and build open and efficient AI platforms deployed at scale. Responsibilities
Work as part of the Accelerator Reference Design Team to specify, develop, and integrate device management for Meta's custom AI hardware Collect requirements and develop specifications for AI/HPC ASIC management and debug functionality Develop and maintain code, to collect, analyze, and interpret data for accelerator fleet statistics Collaborate with cross-functional teams to develop detailed management specifications for our silicon and platforms, focusing on AI accelerator ASIC management Develop concepts and proof-of-concept experiments to improve reliability and visibility of new and existing AI accelerators Specify and Design the module and ASIC management of Meta's custom AI silicon so that it integrates well with the data centers and the fleet Minimum Qualifications
Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience 8+ years of hands-on experience in implementing management solutions for computer server systems 4+ years of designing or specifying custom silicon management Proficiency in C++ and/or Python, or equivalent Experience with general understanding of architectural trade-offs in hardware and software across cost, performance, and reliability Experience working effectively as an individual and in a multidisciplinary team Troubleshooting experience across software, firmware, hardware, and network problems 6+ years of experience in device management for high-performance custom silicon Core domain knowledge in servers and networking with experience in debugging operational problems Datacenter level tooling experience for management Experience in developing both custom hardware and software for AI/HPC Experience with system performance analysis, debug, and optimization practices 10+ years of hands-on experience in implementing device management for computer server systems