Key skills
AI FrameworksPython programmingDistributed computingPerformance tuningGenerative AI techniquesModel parallelismInnovationCollaborationProblem-solvingPythonAIMachine LearningDeep LearningGenerative AILLMPyTorchJAXAnalytics
About this role
NVIDIA is looking for engineers for their core AI Frameworks team to design, develop and optimize real world workloads. The role involves expanding the capabilities of Megatron Core and NeMo Framework, enabling users to develop, train, and optimize models through the implementation of distributed training algorithms and performance tuning.
Responsibilities:
- Develop algorithms for AI/DL, data analytics, machine learning, or scientific computing
- Contribute and advance open source Megatron Core and NeMo Framework
- Solve large-scale, end-to-end AI training and inference challenges, spanning the full model lifecycle from initial orchestration, data pre-processing, running of model training and tuning, to model deployment
- Work at the intersection of compter-architecture, libraries, frameworks, AI applications and the entire software stack
- Innovate and improve model architectures, distributed training algorithms, and model parallel paradigms
- Performance tuning and optimizations, model training and finetuning with mixed precision recipes on next-gen NVIDIA GPU architectures
- Research, prototype, and develop robust and scalable AI tools and pipelines
Requirements:
- MS, PhD or equivalent experience in Computer Science, AI, Applied Math, or related fields and 10+ years of industry experience
- Experience with AI Frameworks (e.g. PyTorch, JAX), and/or inference and deployment environments (e.g. TRTLLM, vLLM, SGLang)
- Proficient in Python programming, software design, debugging, performance analysis, test design and documentation
- 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
- Hands-on experience in large-scale AI training, with a deep understanding of core compute system concepts (such as latency/throughput bottlenecks, pipelining, and multiprocessing) and demonstrated excellence in related performance analysis and tuning
- 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
- Contributions to open source deep learning frameworks