LLM Efficiency Architect: Cross-Layer Performance Leader

NVIDIA Gruppe; Santa Clara (CA); Hybrid; USD 184,000 - 357,000; Full time

Benefits offered by this job

Equity

Benefits

Job summary

NVIDIA is seeking a strong technical leader to drive a unified strategy for making LLMs more efficient from research through deployment. This role blends model innovation, systems expertise, and hardware awareness to ensure capabilities fit compute, memory, power, and cost constraints.

The candidate will lead a multidisciplinary effort, define the direction for LLM efficiency, and influence future model and computing platform designs.

Qualifications

• MS or PhD in Computer Science, Electrical Engineering, Computer Engineering, or related field.
• 5+ years in AI systems, model architecture, computer architecture, HPC, or performance optimization.
• Strong understanding of LLM architectures, training and inference workloads and tradeoffs.
• Strong background in performance analysis, roofline modeling, workload characterization, benchmarking and hardware-aware optimization.
• Proven ability to lead complex optimization projects from concept to production.

Responsibilities

• Lead cross-layer efforts to improve LLM efficiency across model architecture, training and inference systems.
• Analyze LLM workloads mapping to GPUs, memory, interconnects, and distributed infrastructure for co-design opportunities.
• Establish a measurement-driven efficiency roadmap; lead projects from investigation to production deployment.
• Partner with model researchers, systems engineers, compiler/kernel developers, and hardware architects to shape roadmaps.

Skills

AI systems

Model architecture

Computer architecture

High-performance computing

Performance optimization

Technical leadership

Education

MS or PhD degree

Job description

NVIDIA is seeking a strong technical leader to drive a unified strategy for making LLMs more efficient from research through deployment. This role blends model innovation, systems expertise, and hardware awareness to ensure capabilities fit compute, memory, power, and cost constraints.

The candidate will lead a multidisciplinary effort, define the direction for LLM efficiency, and influence future model and computing platform designs.