Senior Performance Architect, Nemotron

Details

We are now looking for a Senior Performance Architect for Nemotron at NVIDIA. NVIDIA is redefining the future of AI systems through deep model–system–hardware co-design. You will shape the next generation of Nemotron models through performance modeling, analysis, and forward projections.

In this role, you will predict before we build—developing high-fidelity models to evaluate how architectural choices translate into real-world deployment efficiency. You will ensure that future models achieve Pareto-optimal trade-offs across accuracy, throughput, and interactivity on target platforms.

Recent efforts such as LatentMoE architectures and the Nemotron Super model exemplify performance-driven co-design—where modeling insights directly shape model architecture and system efficiency at scale. This role sits at the center of Generative AI evolution, partnering across research, framework development, compiler, and hardware teams to guide decisions that determine how efficiently intelligence scales in production.

Responsibilities

• Develop high-fidelity analytical performance models to prototype emerging algorithmic techniques & hardware optimizations to drive model-hardware co-design for the Nemotron family of models.
• Prioritize features to guide future software and hardware roadmap based on detailed performance modeling and analysis.
• Model end-to-end performance impact of emerging GenAI workflows—such as Speculative Decoding, Agentic Pipelines, Inference-time compute scaling, RL, etc.—to understand future datacenter needs.
• Keep up with the latest DL research and collaborate with diverse teams, including DL researchers, hardware architects, and software engineers.

Requirements

• Minimum qualification of a Master’s degree (or equivalent experience) in Computer Science, Electrical Engineering or related fields.
• Strong background in computer architecture, roofline modeling, queuing theory and statistical performance analysis techniques.
• Solid understanding of ML fundamentals, model parallelism and inference serving techniques.
• Proficiency in Python (and optionally C++) for simulator design and data analysis.
• 3+ years of hands-on experience in system evaluation of AI/ML workloads or performance analysis, modeling and optimizations for AI.
• Comfortable defining metrics, designing experiments and visualizing large performance datasets to identify resource bottlenecks.
• Experience with deep learning frameworks like PyTorch, TRT-LLM, VLLM, SGLang.
• A Growth mindset and pragmatic “measure, iterate, deliver” approach.

Benefits

• You will also be eligible for equity and benefits.