Google's TPU Performance team is seeking a Staff Software Engineer to drive performance and efficiency for AI/ML training workloads. This role focuses on Large Language Models, including Google Deepmind Gemini, Bard, and Cloud LLM APIs. The ideal candidate will have extensive experience in software development, machine learning algorithms, and performance optimization. They will work on identifying and maintaining LLM benchmarks, driving TensorFlow/JAX TPU performance, and engaging with product teams to solve LLM performance challenges. The role requires expertise in distributed development, large-scale data processing, and the ability to lead project teams in a complex, matrixed organization. This position offers the opportunity to work on cutting-edge AI technologies and shape the future of machine learning performance at Google.