PhD Internship

Job Description

About Company

ByteDance is dedicated to inspiring creativity and enriching life through its innovative products that help people express themselves, discover, and connect. The company prides itself on its global, diverse teams and a culture that fosters curiosity, humility, and an “Always Day 1” mindset, striving for continuous iteration and meaningful breakthroughs. ByteDance is committed to Diversity & Inclusion, creating an inclusive space that values unique perspectives and reflects the many communities it serves globally. The company also provides reasonable accommodations in its recruitment processes for candidates with disabilities, pregnancy, sincerely held religious beliefs, or other legally protected reasons.

The ByteDance Doubao (Seed) Team, founded in 2023, is at the forefront of pioneering advanced AI foundation models. Its overarching goal is to lead in cutting-edge research and drive significant technological and societal advancements. With a strong commitment to AI, their research areas encompass deep learning, reinforcement learning, Language, Vision, Audio, AI Infra, and AI Safety. The team operates with labs and research positions across China, Singapore, and the US.

Job Description

This job posting is for a PhD Internship position within the Seed Infrastructures team at ByteDance for the 2026 intake.

The Seed Infrastructures team plays a crucial role in overseeing the core technologies for AI foundation models, including:
• Distributed training
• Reinforcement learning framework
• High-performance inference
• Heterogeneous hardware compilation technologies

Internship Overview:
PhD Internships at ByteDance are designed to provide students with a substantial opportunity to actively contribute to the company’s products, research initiatives, future plans, and emerging technologies. The program offers a dynamic experience that combines hands-on learning, enriching community-building and development events, and direct collaboration with industry experts. Applications are reviewed on a rolling basis, and early application is encouraged. Candidates must clearly state their availability (Start date, End date) in their resume.

Summer 2026 Start Dates:
• May 11th, 2026
• May 18th, 2026
• May 26th, 2026
• June 8th, 2026
• June 22nd, 2026

Responsibilities:
The intern will be responsible for a range of critical tasks focused on advancing machine learning systems:
• Research and develop machine learning systems, with a focus on heterogeneous computing architecture, management, scheduling, and monitoring.
• Manage cross-layer optimization of systems, AI algorithms, and hardware (such as GPU, ASIC) specifically for machine learning applications.
• Implement both general-purpose training framework features and model-specific optimizations for advanced models like LLM (Large Language Models) and diffusions.
• Improve efficiency and stability for extremely large-scale distributed training jobs.

Qualifications:

Minimum Qualifications:
Candidates must meet the following criteria:
• Currently enrolled in a PhD program with a strong understanding of distributed and parallel computing principles, as well as recent advancements in computing, storage, networking, and hardware technologies.
• Familiarity with machine learning algorithms, platforms, and frameworks such as PyTorch and Jax.
• A basic understanding of how GPU and/or ASIC works.
• Expertise in at least one or two programming languages commonly used in a Linux environment, specifically C/C++, CUDA, or Python.
• Must obtain and maintain work authorization in the country of employment at the time of hire and throughout employment.

Preferred Qualifications:
The following experiences will be highly advantageous:
• Experience with GPU-based high-performance computing, including RDMA high-performance networks (e.g., MPI, NCCL, ibverbs).
• Familiarity with distributed training framework optimizations such as DeepSpeed, FSDP, Megatron, or GSPMD.
• Knowledge of AI compiler stacks like torch.fx, XLA, and MLIR.
• Experience with large-scale data processing and parallel computing.
• Prior experience in designing and operating large-scale systems in cloud computing or machine learning environments.
• In-depth CUDA programming and performance tuning experience (e.g., using cutlass, triton).

Compensation and Benefits (Campus Intern):
• The hourly rate for this position in the selected city is $85.
• Benefits may vary based on employment nature and country work location. Interns receive day-one access to health insurance, life insurance, and wellbeing benefits.
• Interns are entitled to 10 paid holidays per year and paid sick time (56 hours if hired in the first half of the year, 40 hours if hired in the second half).
• Interns not working 100% remotely may also be eligible for a housing allowance.
• The Company reserves the right to modify or change these benefits programs at any time.

For candidates in Los Angeles County (unincorporated), qualified applicants with arrest or conviction records will be considered in accordance with relevant federal, state, and local laws, including the Los Angeles County Fair Chance Ordinance for Employers and the California Fair Chance Act. The company notes that criminal history may have a direct and negative relationship on job duties involving unsupervised contact with clients/colleagues, handling confidential information, and exercising sound judgment.