Job description

We are now looking for a Senior Performance Architect for Nemotron! At NVIDIA, we are redefining the future of AI systems through deep model–system–hardware co-design. We are looking for a forward-thinking Nemotron Performance Architect to 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 the kind of performance-driven co-design you will help advance—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. What You’ll Be Doing: • Develop high-fidelity analytical performance models to prototype emerging algorithmic techniques & hardware optimizations to drive model-hardware co-design 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 • 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. What we need to see: • A 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. Ways to Stand Out from the Crowd • Proven track record of working in multi-functional teams, spanning algorithms, software and hardware architecture. • Ability to distill complex analyses into clear recommendations for both technical and non-technical collaborators. • Experience with GPU computing (CUDA) Your base salary will be determined based on your location, experience, and the pay of employees in similar positions. The base salary range is 152,000 USD - 241,500 USD for Level 3, and 184,000 USD - 287,500 USD for Level 4. You will also be eligible for equity and benefits . Applications for this job will be accepted at least until May 23, 2026. This posting is for an existing vacancy. NVIDIA uses AI tools in its recruiting processes. NVIDIA is committed to fostering a diverse work environment and proud to be an equal opportunity employer. As we highly value diversity in our current and future employees, we do not discriminate (including in our hiring and promotion practices) on the basis of race, religion, color, national origin, gender, gender expression, sexual orientation, age, marital status, veteran status, disability status or any other characteristic protected by law.

Requirements

Must-haves

Develop high-fidelity analytical performance models to prototype emerging algorithmic techniques & hardware optimizations to drive model-hardware co-design 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
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.
A 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.
Ways to Stand Out from the Crowd
Proven track record of working in multi-functional teams, spanning algorithms, software and hardware architecture.
Ability to distill complex analyses into clear recommendations for both technical and non-technical collaborators.
Experience with GPU computing (CUDA)
You will also be eligible for equity and benefits .
Applications for this job will be accepted at least until May 23, 2026.
This posting is for an existing vacancy.
NVIDIA uses AI tools in its recruiting processes.

Nice-to-haves

None flagged.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aimlpythonllm

Platform: eightfold
Effort: medium