Nvidia Senior LLM Train Framework Engineer 

China, Shanghai 

695391338

What you’ll be doing:

• Build and develop open source Megatron Core.

• Address extensive AI training and inference obstacles, covering the entire model lifecycle including orchestration, data pre-processing, conducting model training and tuning, and deploying models.

• Work at the intersection of AI applications, libraries, frameworks, and the entire software stack.

• Spearhead advancements in model architectures, distributed training strategies, and model parallel approaches.

• Enhance the pace of foundation model training and optimization through mixed precision formulas and advanced NVIDIA GPU structures.

• Performance tuning and optimizations of deep learning framework and software components.

• Research, prototype, and develop robust and scalable AI tools and pipelines.

What we need to see:

• MS, PhD or equivalent experience in Computer Science, AI, Applied Math, or related fields and 5+ years of industry experience.

• Experience with AI train frameworks (e.g., PyTorch, JAX), and/or inference and deployment environments (e.g., TRTLLM, vLLM, SGLang).

• Proficiency in decentralized instruction.

• Proficient in Python programming, software development, debugging, performance analysis, test composition, and documentation.

• CUDA or collective programming skills are a big plus.

• Consistent record of working effectively across multiple engineering initiatives and improving AI libraries with new innovations.

• Strong understanding of AI/Deep-Learning fundamentals and their practical applications.

Ways to stand out from the crowd:

• Proficient in large-scale AI training, knowledgeable in compute system concepts like latency and efficiency.

• Expertise in distributed computing, model parallelism, and mixed precision training.

• Prior experience with Generative AI techniques applied to LLM and Multi-Modal learning (Text, Image, and Video).

• Knowledge of GPU/CPU architecture and related numerical software.

• Familiarity with cloud computing (e.g., complete pipelines for AI training and inference on CSPs like AWS, Azure, GCP, or OCI).