Performance Engineer: High-Scale ML Systems

Anthropic Limited; San Francisco (CA); On-site; USD 280,000 - 850,000; Full time

Job summary

Anthropic is seeking a Performance Engineer to optimize throughput and robustness of our largest distributed systems running at scale. You will identify novel problems, design solutions, and grow into a ML expert while solving challenging systems issues.

The role emphasizes collaboration with researchers, engineers, policy experts, and business leaders to build reliable AI systems that are safe and beneficial for users and society.

Qualifications

• Have significant software engineering or machine learning experience, particularly at supercomputing scale.
• Are results-oriented, with a bias towards flexibility and impact.
• Pick up slack, even if it goes outside your job description.
• Enjoy pair programming.
• Want to learn more about machine learning research.
• Care about the societal impacts of your work.

Responsibilities

• Implement low-latency high-throughput sampling for large language models.
• Implement GPU kernels to adapt our models to low-precision inference.
• Write a custom load-balancing algorithm to optimize serving efficiency.
• Build quantitative models of system performance.
• Design and implement a fault-tolerant distributed system running with a complex network topology.
• Debug kernel-level network latency spikes in a containerized environment.

Skills

Software engineering experience

Machine learning experience

Results-oriented

Pair programming

ML research interest

Societal impact awareness

Flexibility and impact

Education

Bachelor’s degree or equivalent

Tools

GPU/Accelerator programming

ML framework internals

OS internals

Transformers language modeling

Job description

Anthropic is seeking a Performance Engineer to optimize throughput and robustness of our largest distributed systems running at scale. You will identify novel problems, design solutions, and grow into a ML expert while solving challenging systems issues.

The role emphasizes collaboration with researchers, engineers, policy experts, and business leaders to build reliable AI systems that are safe and beneficial for users and society.