AI Computing Software Development Engineer – TensorRT-LLM

About the role

NVIDIA is seeking a Software Development Engineer to join the TensorRT-LLM team, focusing on high‑performance inference for large language models. You will help build the next generation of GPU‑accelerated AI software used across NVIDIA’s product portfolio.

Key responsibilities

• Design, implement, and maintain robust LLM inference software that scales across multiple platforms.
• Perform performance analysis, profiling, and optimization for large‑scale language model inference.
• Stay up‑to‑date with academic and industry advances in AI and integrate relevant updates into TensorRT‑LLM.
• Develop kernels and runtime features to support new LLM architectures and inference algorithms.
• Provide feedback to hardware and architecture teams and collaborate with research, product, and software groups.

Required profile

• Master’s degree or higher in Computer Engineering, Computer Science, Applied Mathematics, or a related field (or equivalent experience).
• At least 3 years of professional software development experience.
• Strong written and oral communication skills in English.
• Proactive, able to work independently, and comfortable in a fast‑paced delivery environment.

Required skills

• Excellent Python programming and software design.
• Experience with deep‑learning frameworks such as PyTorch and HuggingFace.
• Strong C/C++ development, debugging, and test design.
• GPU programming expertise (CUDA or OpenCL) and architectural knowledge of CPU/GPU systems.
• Performance modeling, profiling, and optimization of high‑performance applications.
• Familiarity with LLM inference workflows and frameworks (e.g., TensorRT‑LLM, SGLang, vLLM).

What we offer

• Opportunity to work on cutting‑edge AI and compute technologies.
• Exposure to the full deep‑learning software stack.
• Competitive salary and generous benefits package.