Technology · Shanghai, China

Full-time

Software Engineering Intern, NCCL - 2027

NVIDIA · Shanghai, China · Posted Nov 13, 2026

This posting comes from NVIDIA. Confirm the details on their site before applying.

Overview

NVIDIA is leading the way in groundbreaking developments in Artificial Intelligence, High-Performance Computing and Visualization. The GPU, our invention, serves as the visual cortex of modern computers and is at the heart of our products and services. Our work opens up new universes to explore, enables amazing creativity and discovery, and powers what were once science fiction inventions from artificial intelligence to autonomous cars.

Responsibilities

  • Design, implement and maintain highly-optimized communication runtimes for Deep Learning frameworks (e.g. NCCL for TensorFlow/Pytorch) and HPC programming interfaces (e.g. UCX for MPI/OpenSHMEM) on GPU clusters.
  • Participating in and contributing to parallel programming interface specifications like MPI/OpenSHMEM.
  • Design, implement and maintain system software that enables interactions among GPUs and interactions between GPUs and other system components.
  • Creating proof-of-concepts to evaluate and motivate extensions in programming models, new designs in runtimes and new features in hardware.
  • You are pursuing a Ph.D. in CE/CS/EE with a strong background in computer architecture, operating systems, communication library and/or AI/ML.
  • Excellent C/C++ programming and debugging skills.
  • Strong experience with Linux.
  • Experience with parallel programming interfaces and communication runtimes.
  • Ability and flexibility to work and communicate effectively in a multi-national, multi-time-zone corporate environment.

Preferred qualifications

  • Deep knowledge of high-performance networks like InfiniBand, RoCE etc.
  • Background with HPC applications. Experience with Deep Learning Frameworks such PyTorch, TensorFlow, JAX/XLA, vLLM/SGLang etc.
  • Experience with AI/DL communication patterns such as Expert Parallelism (EP), TP, DP, PP and how these patterns can be implemented with NCCL. Experience with CUDA kernel optimization and profiling.
  • Experience with large-scale model training and production inference software stack.
  • Strong collaborative and interpersonal skills, specifically a proven ability to effectively guide and influence within a dynamic matrix environment.

Great fits start here.

Whether you're looking for your next role or your next team member, give people and possibilities a place to meet.