PhD Research Intern: RDMA & High-Speed Networking

ByteDance; San Jose (CA); On-site; USD 68,191 - 97,120; Part time

Job summary

ByteDance is hiring a Research Intern (RDMA/High Speed Network) in San Jose to start in 2026. You will design, optimize and deploy high-performance transport protocols and congestion control for AI/LLM applications, and explore AI network frameworks and protocols.

The role requires pursuing a PhD in computer networking, strong programming in C/C++, Python or Go, and knowledge of RDMA and high-speed networks. This internship offers hands-on research with real impact.

Qualifications

• Currently pursuing a PhD in computer networking or a related field.
• Experience with TCP, RoCEv2 and socket/verbs based network programming.
• Familiar with data center congestion control algorithms.
• Knowledge of scale-up protocols like PCIe, NVLink, UALink and scale-out networks.
• Proficiency in C/C++, Python, or Go.
• Knowledge of GPU architecture.
• Experience with NCCL, MPI or RPC libraries is a plus.

Responsibilities

• Design, optimize, implement and deploy high-performance transport protocols for AI/LLM apps.
• Design and optimize congestion control algorithms for AI workloads.
• R&D of high-performance AI communication frameworks and protocol stacks.
• Survey academia and industry advances and present findings in papers.

Skills

C/C++

Python

Socket programming

RDMA

High-speed networks

Go

GPU architecture

NCCL

MPI / RPC

Education

PhD in computer networking

Tools

NCCL

MPI

RPC libraries

Job description

ByteDance is hiring a Research Intern (RDMA/High Speed Network) in San Jose to start in 2026. You will design, optimize and deploy high-performance transport protocols and congestion control for AI/LLM applications, and explore AI network frameworks and protocols.

The role requires pursuing a PhD in computer networking, strong programming in C/C++, Python or Go, and knowledge of RDMA and high-speed networks. This internship offers hands-on research with real impact.