NVIDIA has been transforming computer graphics, PC gaming, and accelerated computing for more than 25 years. It’s a unique legacy of innovation that’s fueled by great technology and amazing people. Today, we’re tapping into the unlimited potential of AI to define the next era of computing. An era in which our GPU acts as the brains of computers, robots, and self-driving cars that can understand the world. Doing what’s never been done before takes vision, innovation, and the world’s best talent. As an NVIDIAN, you’ll be immersed in a diverse, supportive environment where everyone is inspired to do their best work. Come join the team and see how you can make a lasting impact on the world.

What You’ll be doing:

As an intern, you’ll focus on one of three specialized tracks:

• TensorRT LLM – Inference Optimization (Python / PyTorch): Build and enhance high‑performance LLM inference pipelines. Analyze and optimize model execution, scalability, and memory usage. Collaborate across framework and research teams to deliver efficient multi‑GPU model serving.

• TensorRT Compiler – Graph Optimization (C++): Work on the TensorRT compiler backend to improve graph transformations and code generation for NVIDIA GPUs. Develop compiler optimization passes, refine operator fusion and memory allocation, and collaborate with CUDA and hardware architecture teams.

• CuTe DSL & CUDA Kernels Development / Optimization (CUDA C++ / Python DSL): Design and tune GPU compute kernels and DSL implementations for core deep learning operations such as GEMM, MoE, Attention and Convolution. Profile, analyze and improve CUDA kernel performance to achieve maximum GPU efficiency.

What We need to see:

• Pursuing an M.S. or Ph.D. in Computer Science, Computer Engineering, Electrical Engineering, Applied Mathematics, or related field.

• Excellent problem‑solving ability, curiosity for cutting‑edge AI systems, and passion for GPU computing and deep learning software performance.

• TensorRT LLM: Strong Python programming and experience with PyTorch; solid understanding of inference and GPU acceleration.

• TensorRT Compiler: Proficient in C++, with experience in compiler or performance optimization.

• CuTe DSL & CUDA Kernels: Skilled in C/C++ and CUDA or parallel programming; familiar with LLVM, MLIR and compiler; understanding of computer architecture and performance profiling/analysis/optimization.

Join us and play a part in building the AI computing platforms that drive innovation across industries worldwide.