Deep Learning Software Engineer, Inference - New College Grad 2026

About the role

NVIDIA is seeking a Software Engineer specializing in Deep Learning Inference to join a dynamic team focused on advancing the frontiers of AI. This role involves designing, building, and optimizing GPU-accelerated software that powers sophisticated AI applications.

Responsibilities

• Performance optimization, analysis, and tuning of DL models in various domains like LLM, Multimodal and Generative AI.
• Scale performance of DL models across different architectures and types of NVIDIA accelerators.
• Contribute features and code to NVIDIA’s inference libraries, vLLM and SGLang, FlashInfer and LLM software solutions.
• Work with cross-collaborative teams across frameworks, NVIDIA libraries and inference optimization innovative solutions.

Requirements

• Pursuing or recently completed a MS or PhD in Computer Engineering, Computer Science, EECS, AI or related field or equivalent experience.
• Software development experience.
• Excellent C/C++ programming and software design skills.
• SW Agile skills are helpful and Python experience is a plus.
• Prior experience with training, deploying or optimizing the inference of DL models in production is a plus.
• Prior background with performance modeling, profiling, debug, and code optimization or architectural knowledge of CPU and GPU is a plus.
• GPU programming experience (CUDA, OAI TRITON or CUTLASS) is a plus.

Qualifications

• Experience with deep learning software projects, such as PyTorch, vLLM, and SGLang to drive advancements in the field.
• Experience with Multi GPU Communications (NCCL, NVSHMEM).

Skills

• Open-source tools and plugins including CUTLASS, OAI Triton, NCCL, and CUDA kernels.

Benefits

• Competitive base salary ranging from $124,000 to $195,500 for Level 2 and $152,000 to $241,500 for Level 3.
• Eligibility for equity and benefits.

Pay

Base salary range: $124,000 - $195,500 for Level 2, and $152,000 - $241,500 for Level 3.

Schedule

Not specified.

Benefits

• Not specified.