Inference Optimization Intern: GPU Performance Modeling

Ifm Us; Sunnyvale (CA); On-site; USD 30,000 - 60,000; Full time

Job summary

Ifm Us is seeking an intern to work on the optimization of large-scale foundation models using NVIDIA GPU architectures. Interns will have the chance to collaborate with world-class researchers and engineers, gaining hands-on experience in GPU performance analysis and kernel optimization.

The ideal candidate should be pursuing a degree in Computer Science or related fields, with strong programming skills in CUDA, C++, and Python. This is a unique opportunity to contribute to cutting-edge developments in AI systems.

Qualifications

• Currently pursuing a degree in a relevant quantitative discipline.
• Experience with CUDA programming and GPU kernel development.
• Understanding of NVIDIA GPU architecture and memory hierarchy.

Responsibilities

• Develop analytical performance models for GPU kernels.
• Build and validate a simulator for performance estimation.
• Identify performance bottlenecks in compute and memory.

Skills

CUDA programming

GPU kernel development

C++

Python

Nsight Systems

Nsight Compute

Deep learning frameworks (PyTorch, TensorFlow)

Performance analysis

Education

Pursuing degree in Computer Science or related field

Job description

Ifm Us is seeking an intern to work on the optimization of large-scale foundation models using NVIDIA GPU architectures. Interns will have the chance to collaborate with world-class researchers and engineers, gaining hands-on experience in GPU performance analysis and kernel optimization.

The ideal candidate should be pursuing a degree in Computer Science or related fields, with strong programming skills in CUDA, C++, and Python. This is a unique opportunity to contribute to cutting-edge developments in AI systems.