TPU Kernel Engineer

Overview

TPU Kernel Engineer responsible for diagnosing ML performance issues across research, training, and inference, and for designing/optimizing TPU kernels.

What you'll do

• Identify and address performance issues across multiple ML systems.
• Design and optimize kernels for TPU.
• Provide feedback to researchers on how model changes affect performance.
• Implement low-latency, high-throughput sampling for large language models.
• Adapt models for low-precision inference.
• Debug kernel performance at the assembly level.

What you'll need

• Significant experience optimizing ML systems for TPUs, GPUs, or other accelerators.
• Track record solving large-scale systems problems.
• Low-level optimization experience.
• Bias towards flexibility and impact.
• Ability to pick up work outside the job description.
• Experience with pair programming.

Details

• Annual compensation range: $280,000–$850,000 USD.
• Minimum education: Bachelor’s degree or an equivalent combination of education, training, and/or experience.
• Location-based hybrid policy: expected to be in one of the offices at least 25% of the time.
• Visa sponsorship: company sponsors visas for some roles/candidates and makes reasonable efforts if an offer is made.