TPU Kernel Engineer

Anthropic
San Francisco, New York, Seattle
Workplace: OnsiteFull timeUSD 350,000 - 850,000 annuallyFunction: Healthcare (Clinical, Medical, Wellness)Education: bachelorsSkills: ["Assembly language","Kernel development","Parallel programming","Low-level optimization","High-performance computing"]

As a TPU Kernel Engineer, you’ll identify and address performance issues across ML systems (research, training, and inference) and design/optimize kernels for TPUs. You’ll provide feedback to researchers on how model changes impact performance, tackle large-scale systems problems, and work on low-level optimization—often collaborating in a highly technical, research-driven environment.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Anthropic
Anthropic
1 year ago

TPU Kernel Engineer

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 12 hours agoStatus: Live

Job Summary

As a TPU Kernel Engineer, you’ll identify and address performance issues across ML systems (research, training, and inference) and design/optimize kernels for TPUs. You’ll provide feedback to researchers on how model changes impact performance, tackle large-scale systems problems, and work on low-level optimization—often collaborating in a highly technical, research-driven environment.
Location: San Francisco, New York, Seattle
Workplace: Onsite
Employment Type: Full time
Job Function: Healthcare (Clinical, Medical, Wellness)

Key Responsibilities

  • •Identify and address performance issues across ML systems (research, training, inference) and design/optimize kernels for the TPU
  • •Provide feedback to researchers on how model changes impact performance and scalability
  • •Tackle large-scale systems problems and pursue low-level optimizations to improve throughput and latency
  • •Collaborate with researchers and engineers; participate in pair programming and code reviews
  • •Develop and debug kernel performance improvements, including potential assembly-level analysis and optimization

Pay and Benefits

Salary: USD 350,000 - 850,000 annually

Key Requirements

  • •Significant experience optimizing ML systems for TPUs, GPUs, or other accelerators; ability to identify performance bottlenecks and implement impactful improvements
  • •Strong teamwork and communication skills; willingness to pair program and collaborate closely with researchers
  • •Experience with low-level optimization and understanding of computer architecture; ability to reason about kernel performance at the assembly level
  • •Background in ML model development or ML framework internals; familiarity with transformers or large language models is a plus
  • •Bachelor's degree in a related field or equivalent experience
Experience:AIMachine learningAccelerators
Education:Bachelor's
Skills:Assembly languageKernel developmentParallel programmingLow-level optimizationHigh-performance computing
Languages:English
Tech Stack:TPUGPUAcceleratorsML frameworks

Company Brief

Anthropic
Develops large-scale AI systems and safety research to create reliable, steerable, and interpretable AI assistants and models for commercial and research applications.
Industry: AI & Machine Learning
Company Size: Enterprise (1,001+ employees)
Growth: Scaleup
Valuation: Unicorn (USD 1B+)
Funding: Series C
Headquarters: San Francisco, United States
Founded: 2021
WebsiteLinkedIn