Job Summary:
NVIDIA is redefining the future of AI systems through deep modelโsystemโhardware co-design, and they are looking for a Senior Performance Architect for Nemotron. In this role, you will develop high-fidelity performance models to evaluate architectural choices and ensure future models achieve optimal trade-offs across accuracy and efficiency.
Responsibilities:
โข Develop high-fidelity analytical performance models to prototype emerging algorithmic techniques & hardware optimizations to drive model-hardware co-design Nemotron family of models.
โข Prioritize features to guide future software and hardware roadmap based on detailed performance modeling and analysis
โข Model end-to-end performance impact of emerging GenAI workflows - such as Speculative Decoding, Agentic Pipelines, Inference-time compute scaling, RL etc. โ to understand future datacenter needs
โข This position requires you to keep up with the latest DL research and collaborate with diverse teams, including DL researchers, hardware architects, and software engineers.
Qualifications:
Required:
โข A minimum qualification of a Master's degree (or equivalent experience) in Computer Science, Electrical Engineering or related fields.
โข Strong background in computer architecture, roofline modeling, queuing theory and statistical performance analysis techniques.
โข Solid understanding of ML fundamentals, model parallelism and inference serving techniques.
โข Proficiency in Python (and optionally C++) for simulator design and data analysis.
โข 3+ years of hands-on experience in system evaluation of AI/ML workloads or performance analysis, modeling and optimizations for AI.
โข Comfortable defining metrics, designing experiments and visualizing large performance datasets to identify resource bottlenecks.
โข Experience with deep learning frameworks like PyTorch, TRT-LLM, VLLM, SGLang.
โข A Growth mindset and pragmatic 'measure, iterate, deliver' approach.
Preferred:
โข Proven track record of working in multi-functional teams, spanning algorithms, software and hardware architecture.
โข Ability to distill complex analyses into clear recommendations for both technical and non-technical collaborators.
โข Experience with GPU computing (CUDA)
Company:
NVIDIA is a computing platform company operating at the intersection of graphics, HPC, and AI. Founded in 1993, the company is headquartered in Santa Clara, USA, with a team of 10001+ employees. The company is currently Late Stage.