PhD Intern – AI/LLM Networking

Job Description

Research Intern (AI/LLM Network) – 2026 Start (PhD) | ByteDance

The Tone:
This is an internship at ByteDance, with a 2026 start date for PhD students. ByteDance builds hyperscale data-center networking solutions that power popular global applications like TikTok and Douyin, serving hundreds of millions of users. This role is crucial for designing, building, and operating the intelligent network infrastructure required for high availability, scalability, and performance, specifically tailored for AI/LLM applications. The intern will gain marketable software development and network operation experience at a massive scale.

The TL;DR
• Role: Internship
• Pay: $60 hourly
• Team: ByteDance Networking team is responsible for hyperscale data-center networking solutions.
• Mission: This person will contribute to the design, implementation, and optimization of high-speed network technologies and systems to enhance the scalability, reliability, and performance of AI/LLM networks.
• Tech Stack: C/C++, Python, Go, RDMA, NCCL, MPI, RPC libraries

What You’ll Actually Do
• Design: Design, implement, and deploy high-speed network technologies to support AI/LLM applications.
• Develop: Design and develop platforms and systems for monitoring, analysis, and diagnosis of large-scale AI/LLM networks.
• Research: Research and develop high-performance AI communication frameworks, network protocol stacks, and co-design optimizations of host-network-application components.
• Innovate: Follow the latest technologies from academia and industry, identify innovative system parts, and present findings in academic papers.

The Must-Haves
• Background: Currently pursuing a PhD in computer networking or a related technical discipline.
• Experience: Developing high-performance communication frameworks (including NCCL, MPI, and RPC libraries) or developing software systems for AI network diagnosis and performance optimization.
• Skills: Proficiency in computer network and network programming; proficiency in C/C++, Python, or Go; familiarity with high-speed network systems including RDMA, congestion control, and AI network optimization.
• Bonus: Familiarity with AI training/inference systems and software-hardware co-design; having top-tier networking conference publications such as NSDI, SIGCOMM, OSDI, or SOSP.