GenAI Inference Optimization Intern – CUDA & GPUs

NVIDIA Corporation; Polska; On-site; PLN 118,000 - 204,000; Full time

Benefits offered by this job

Comprehensive benefits

Equal opportunity employer

Job summary

NVIDIA Corporation seeks a GenAI Algorithms Intern to design, implement, and productionize model optimization algorithms for inference on NVIDIA hardware. You will work on quantization, pruning, and distillation across LLMs, VLMs, and diffusion models to improve speed and memory efficiency.

You'll contribute to NVIDIA's Model Optimizer ecosystem, collaborate with teams across NeMo/Megatron and TensorRT-LLM, and gain hands-on experience with CUDA, Triton, and PyTorch in a fast-paced research

Qualifications

• Pursuing a Master's or PhD in Computer Science, AI, Applied Math, or a related field.
• Strong deep learning background and hands-on experience with language models.
• Solid foundation in algorithms and programming.
• Proficiency in Python, PyTorch, and modern ML frameworks/tools.
• Good written and verbal communication skills, with the ability to work independently and collaboratively.

Responsibilities

• Design and build modular, scalable model optimization software platforms for diverse AI models.
• Explore, develop, and integrate deep learning optimization algorithms into NVIDIA's AI stack.
• Deploy optimized models into leading open-source inference frameworks and contribute APIs and features.
• Collaborate with NVIDIA teams on model optimization solutions for customer use cases.
• Conduct GPU kernel-level profiling to identify optimization opportunities.

Skills

Deep learning background

Python

PyTorch

Communication skills

Education

Master's or PhD candidate in Computer Science/AI/Applied Math

Tools

CUDA

Triton

NeMo

Megatron

PyTorch

Job description

NVIDIA Corporation seeks a GenAI Algorithms Intern to design, implement, and productionize model optimization algorithms for inference on NVIDIA hardware. You will work on quantization, pruning, and distillation across LLMs, VLMs, and diffusion models to improve speed and memory efficiency.

You'll contribute to NVIDIA's Model Optimizer ecosystem, collaborate with teams across NeMo/Megatron and TensorRT-LLM, and gain hands-on experience with CUDA, Triton, and PyTorch in a fast-paced research