Google Cloud is seeking a Silicon Infrastructure Engineer specializing in Emulation to join their team. This role is crucial in shaping the future of AI/ML hardware acceleration, particularly working with TPU (Tensor Processing Unit) technology that powers Google's most demanding AI/ML applications. The position involves developing and maintaining emulation infrastructure, tools, and workflow methodologies supporting critical ASIC projects.
The role sits within the ML, Systems, & Cloud AI (MSCA) organization, which is responsible for designing, implementing, and managing hardware, software, machine learning, and systems infrastructure for all Google services and Google Cloud. You'll work directly with emulation team members, designers, verification engineers, and software teams, while also interfacing with external vendors, lab support teams, and EDA tooling teams.
Key responsibilities include maintaining emulation infrastructure, implementing new methodologies, creating automation tools, and supporting debug efforts. The ideal candidate should have strong experience in electrical engineering, emulation systems, and programming skills. This position offers the opportunity to work on cutting-edge technology that powers Google's AI/ML capabilities and impacts billions of users worldwide.
The role requires a blend of technical expertise in hardware emulation, software development, and system administration. You'll be part of a team that prioritizes security, efficiency, and reliability while driving towards shaping the future of hyperscale computing. This is an excellent opportunity for someone passionate about hardware infrastructure and interested in working at the intersection of silicon development and AI/ML technologies.