Generative AI Accelerator Architect and Performance Architect
San DiegoLast seen today
Summary
Design and optimize custom silicon solutions for Google's generative AI mobile processors (Tensor SoCs). Drive ML architecture exploration for LLMs, Vision Transformers, and diffusion models across the full hardware-software stack, collaborating with research, system architecture, and compiler teams. Define system requirements, apply advanced research in architecture and process technology, and optimize GenAI workload scheduling on TPU compute engines. Requires 10+ years in chip/computer architecture, performance engineering, or compilers, plus hands-on experience with modern generative AI model types and deep learning frameworks.
Data ScienceEmbedded SystemsAIC++Machine LearningPythonChip ArchitectureCompiler DesignDesign Space ExplorationGpuHardware Software Co DesignImage Diffusion ModelsJaxLLM Large Language ModelsNpuPerformance OptimizationPower AnalysisPyTorchSystem ArchitectureSystem Pnp ModelingTensorFlowThermal AnalysisTpuVision Transformers