NVIDIA Deep Learning Architecture Intern
Apply for this position at this link and give us a feedback answering the questions at the end of this Gomry form.
The NVIDIA Architecture group is looking for world-class Interns in Architecture to join and lead their various efforts.
This is a full time internship and it will take place in Santa Clara, CA, US.
A fast growing direction in the group is to study performance of large-scale generative AI models and neural graphics for gaming and physical simulation in order to deliver the highest full system performance in the world.
NVIDIA is constantly looking for ways to improve their architecture and maintain their leadership by developing new parallel programming models, new architectures and infrastructure that is required to make this successful.
What you will be doing:
Research and performance study large-scale generative AI models such as LLM and Stable Diffusion.
Research and performance study neural graphics for gaming and physical simulation.
Working with other architects to identify shortcomings of GPU and recommend improvement.
Performance profiling tool development.
Requirements:
Pursuing your MS or PHD in EE/CS/EECS or related technical field. Ideally passionate about Deep Learning algorithms and Computer Architecture.
Meaningful experience with Deep Learning application development and GPU performance profiling. Knowledge of CUDA or DirectML is preferred.
Strong programming ability in C and C++. Python scripting a big plus.
Proven background in mathematics, algorithms and data structures required, additional background in 3D Compute Graphics preferred.
The hourly rate for our interns is $19 - $93. The internship hourly rates are a standard pay determined based on the position and your location, year in school, degree, and experience.
You will also be eligible for Intern benefits.
Learn more and apply here.