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.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
Job Type
Full-time
Career Level
Intern
Education Level
Ph.D. or professional degree