About the role#
The PyTorch TPU team works on running large models efficiently on TPU hardware. This internship involves performance modeling, parallelism design, and kernel development. Internships last between 12 and 24 weeks with various start dates available throughout the year.
What you'll do#
- Build and test systems to run LLM and multi-modality models on TPU hardware at a lower cost.
- Study traffic patterns to find computational insights.
- Apply these insights to improve system and kernel performance.
What you'll need#
- You are currently pursuing or have completed a PhD in Computer Science, Computer Engineering, Electrical Engineering, or a related field.
- You have professional experience with Python.
- You have worked on system and kernel optimization.
- You must have and maintain work authorization in the United States.
- Experience with GPU, TPU, or ASIC hardware is a plus.
- Familiarity with Triton, Cuda, Cutedsl, Pallas, or performance modeling is preferred.
Location & details#
- This role is located in Menlo Park, California.
- This is a full-time, paid, on-site internship.
- Compensation ranges from $7,650 to $12,667 per month.
About Meta
Meta develops software and social media applications. It operates platforms such as Facebook, Messenger, Instagram, and WhatsApp. The company maintains headquarters in Menlo Park, California. Current work focuses on augmented and virtual reality alongside artificial intelligence and machine learning.
How to get in at Meta
Landing an internship at Meta requires a proactive approach to the application process. Intern Insider sends an instant alert the moment a role matching your target is published, allowing you to apply among the first. Early applicants often have their materials reviewed before the applicant pool grows significantly. You can use Intern Insider to surface the recruiters behind specific roles at Meta. Reaching out directly to these individuals to ask about the position or a referral often improves your response rates compared to applying into a queue.
