This role involves driving forward-looking Generative AI (GenAI) Machine Learning architecture exploration for Tensor mobile SoCs. Collaboration with research, system architecture, and compiler teams is crucial to optimize future workloads across the entire tech stack (hardware, software, use cases, network, and external components). Responsibilities include defining system architecture requirements for future GenAI use cases, applying advanced research to achieve power and performance improvements on GenAI workloads, and optimizing GenAI use case performance through model scheduling on TPU compute engines. The ideal candidate possesses extensive experience in computer architecture, performance, and compilers, coupled with expertise in Generative AI model architectures (LLMs, Vision Transformers, etc.). Proficiency in programming languages (C/C++, Python) and deep learning frameworks (TensorFlow/Jax/PyTorch) is essential.