Deep Learning Intern

What you'll do

• Develop algorithms for deep learning, data analytics, and scientific computing to improve GPU performance.
• Build underlying frameworks and libraries to accelerate deep learning on GPUs.
• Contribute directly to major software packages such as JAX, PyTorch, and TensorFlow.
• Integrate the latest library features like cuDNN or CUDA into existing software.
• Perform performance tuning and analysis for deep learning algorithms and libraries.
• Maintain build, test, and distribution infrastructure for libraries across NVIDIA supported platforms.
• Optimize core deep learning libraries such as CuBLAS and CuDNN.

What we're looking for

• Must be actively enrolled in a university pursuing a B.S., M.S., or Ph.D. degree in Electrical Engineering, Computer Engineering, or a related field.
• Anticipated graduation date must be clearly indicated on the resume or CV.
• Experience with Deep Neural Networks, Linear Algebra, Numerical Methods, and/or Computer Vision may be required.
• Experience with Software Design, Computer Memory, CPU and GPU Architectures, Networking, and Embedded System Design may be required.
• Experience with Driver development and Real-Time Software may be required.
• Experience with Computer Architecture (CPUs, GPUs, FPGAs), GPU Programming Models, and Performance-Oriented Parallel Programming may be required.
• Knowledge of programming skills and technologies such as C, C++, CUDA, Python, x86, ARM CPU, GPU, Linux, and Bash/Shell Scripting may be required.
• Experience with Graphics APIs (Direct3D, Vulkan, OpenGL), OpenCL, or container tools like Docker and Kubernetes may be required.