Research Scientist Intern, TPU(PhD)

Meta•Menlo Park, CA
•$7,650 - $12,667

About The Position

The PyTorch TPU team focuses on enabling efficient and accurate inference of first-party models directly on TPU. The work spans performance modeling, sharding and parallelism design, kernel authoring, and automation with agents. This internship lasts 12 to 24 weeks, with various start dates throughout the year. The intern will develop and evaluate methodologies and systems for running large language and multimodality models cost-effectively on TPU systems. They will analyze traffic patterns and extract insights to derive computational intrinsics that can inform system and kernel optimization.

Requirements

  • Currently has or is in the process of obtaining a PhD in computer science, computer engineering, electrical engineering, or a related field.
  • Must obtain work authorization in the country of employment by the time of hire and maintain ongoing work authorization during employment.
  • Proficient in Python.
  • Experience with system and kernel optimization.

Nice To Haves

  • Experience with GPUs, TPUs, or ASICs.
  • Experience with Triton, CUDA, CuteDSL, or Pallas.
  • Familiarity with performance modeling.

Responsibilities

  • Develop and evaluate methodologies and systems for running LLM and multimodality models cost-effectively on TPU systems.
  • Analyze traffic patterns and extract insights to derive computational intrinsics that can be used to optimize systems and kernels.

Benefits

  • Benefits are mentioned, but no specific benefits are listed.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service