TPU Kernel Engineer for High-Performance ML Systems

SignalAI; New York (NY); Hybrid; USD 280,000 - 850,000; Full time

Job summary

Anthropic is seeking a TPU Kernel Engineer to identify and address performance issues across ML systems, including research, training, and inference. You will design and optimize kernels for the TPU and provide feedback to researchers about how model changes impact performance.

Strong candidates will have a track record of solving large-scale systems problems and low-level optimization, with experience on TPUs, GPUs, or accelerators, and an interest in ML research and societal impact.

Qualifications

• Experience optimizing ML systems for TPUs, GPUs, or accelerators.
• Bias towards flexibility and impact in work.
• Willing to take on tasks outside job description.
• Enjoy pair programming and collaboration.
• Interest in machine learning research.
• Concern for societal impacts of AI work.

Responsibilities

• Implement low-latency, high-throughput sampling for large language models.
• Adapt existing models for low-precision inference.
• Build quantitative models of system performance.
• Design and implement custom collective communication algorithms.
• Debug kernel performance at the assembly level.

Skills

TPU optimization

Results oriented

Pair programming

ML research interest

Societal impact awareness

Education

Bachelor's degree

Tools

Low-level kernel optimization

ML framework internals

Transformer architectures

Computer architecture

Job description

Anthropic is seeking a TPU Kernel Engineer to identify and address performance issues across ML systems, including research, training, and inference. You will design and optimize kernels for the TPU and provide feedback to researchers about how model changes impact performance.

Strong candidates will have a track record of solving large-scale systems problems and low-level optimization, with experience on TPUs, GPUs, or accelerators, and an interest in ML research and societal impact.