The Staff Software Engineer, ML Compiler Performance Engineer Co-Design role at Google focuses on enhancing machine learning workloads to achieve optimal performance on TPU hardware. The position is part of the MARC team, which operates at the intersection of ML workloads and TPU silicon roadmaps. The engineer will define strategies for integrating open-source ML models and ensure they meet performance benchmarks. This role requires collaboration with technical leadership across various teams to drive multi-year roadmaps.

• Bachelor's degree or equivalent practical experience.
• 8 years of experience programming in C++ or Python.
• 5 years of experience testing and launching software products.
• 5 years of experience with performance, large-scale systems data analysis, visualization tools, or debugging.
• 3 years of experience with software design and architecture.

• Deliver certified performance benchmarks six months pre-pilot as primary targets for pricing, capacity planning, and kernel optimization.
• Adapt emerging open-source models into TPU-native paired variants, including sparsity optimization, to demonstrate differentiated TPU performance.
• Ingest and optimize multi-turn agentic workflows, reasoning loops, and prompt and decode topologies prior to silicon tape-out.
• Author robust reference implementations using PyTorch, JAX, or Pallas that meet pre-pilot goals under compiler constraints.
• Partner with director and executive technical leadership across TPU Hardware, XLA Compiler, and Cloud AI Systems to steer multi-year roadmaps.

• Employees at Google are often offered benefits like comprehensive health insurance, 401(k) matching, and flexible work arrangements, among other benefits.