Skip to main content
NVIDIA logo

Performance Software Intern, Deep Learning Libraries - 2027

NVIDIA
September 25, 2026
Internship
On-site
Shanghai, China
EDA Jobs, Level - Entry or Early Career

Job Title

Performance Software Intern, Deep Learning Libraries - 2027

Role Summary

Internship role on the deep learning libraries performance team responsible for developing and optimizing code that accelerates linear algebra and deep learning operations on NVIDIA GPUs. Work focuses on low-level software close to GPU hardware to maximize throughput and efficiency.

Experience Level

Entry-level internship intended for early-career candidates (internship position).

Responsibilities

Primary responsibilities for this internship include implementing and tuning compute kernels, validating performance, and collaborating with cross-functional teams.

  • Write highly tuned compute kernels for core deep learning operations (e.g., matrix multiply, MoE, Attention).
  • Perform performance analysis, identify bottlenecks, and optimize resource utilization to improve throughput.
  • Follow software engineering best practices, including regression testing and CI/CD integration.
  • Collaborate with compiler, deep learning performance, and hardware architecture teams to generate optimal code and use new hardware features.

Requirements

Must-have technical skills and experience for successful contribution; nice-to-have items listed separately.

  • Must-have: Strong programming and software design skills, including debugging, performance analysis, and test design.
  • Must-have: Experience with performance-oriented parallel programming (e.g., OpenMP, pthreads, or equivalent).
  • Must-have: Solid understanding of computer architecture and some experience with assembly programming.
  • Must-have: Proven ability to find performance bottlenecks and implement optimizations.
  • Nice-to-have: Prior tuning of deep learning library kernel code.
  • Nice-to-have: CUDA GPU programming experience.
  • Nice-to-have: Background in numerical methods and linear algebra.
  • Nice-to-have: Experience with LLVM, TVM tensor expressions, or TensorFlow MLIR.

Education Requirements

Pursuing a Master’s or PhD in Computer Science, Computer Engineering, Applied Mathematics, or a related technical field (graduate student status expected).


About the Company

Company: NVIDIA

Headquarters: Santa Clara, California, USA

NVIDIA is a global leader in accelerated computing, renowned for its innovative solutions in AI and digital twins that transform diverse industries. The company specializes in networking technologies, providing end-to-end InfiniBand and Ethernet solutions for servers and storage that optimize performance and scalability. NVIDIA serves sectors such as high-performance computing, enterprise data centers, and cloud computing, constantly reinventing its products and services to stay ahead in the market.

NVIDIA logo

Date Posted: 2026-09-23