Embedded Systems Engineer, High Speed DSP

In this role, you’ll work to shape the future of AI/ML hardware acceleration. You will have an opportunity to drive cutting-edge TPU (Tensor Processing Unit) technology that powers Google's most demanding AI/ML applications. You’ll be part of a team that pushes boundaries, developing custom silicon solutions that power the future of Google's TPU. You'll contribute to the innovation behind products loved by millions worldwide, and leverage your design and verification expertise to verify complex digital designs, with a specific focus on TPU architecture and its integration within AI/ML-driven systems. In this role, for High-Speed DSP, you will architect and define the low-level firmware and control loops powering Google's physical-layer communications silicon. Sitting at the critical intersection of digital signal processing (DSP), hardware-software co-design, and real-time systems, you will lead the development of adaptive algorithms, error correction (FEC), and link-telemetry frameworks. Your work will directly bridge the gap between high-performance hardware pipelines and Google’s AI infrastructure, transforming complex communication systems theory into reliable, real-time silicon execution.

Define the long-term firmware roadmap and hardware-software partitioning for physical-layer interfaces, determining the optimal balance between hardware-fixed logic and programmable firmware."urlArchitect advanced "in-situ" monitoring systems and firmware-driven diagnostics to track physical-layer link health, Bit Error Rate (BER) trends, and channel quality across millions of ports in real-time."urlDesign, model, and implement bare-metal or RTOS firmware to control high-speed transceivers, managing real-time adaptive loops, clock and data recovery (CDR), and physical-layer calibration."urlPartner with TPU Systems, Networking Software, and High-Performance Computing (HPC) teams to ensure silicon firmware integrates seamlessly with global cluster management and collective communication frameworks."urlIdentify and de-risk hardware-software interface bottlenecks (e.g., register maps, driver latencies, and boot-up sequencing) that impact Google's AI infrastructure.